Replicating "The Structure of Economic News" (Bybee et al. 2020)

  • author: Tao Wang
  • This notebook builds on the companion notebook LDA-example and extends it to closely follow (Bybee et al. 2020)'s implementation of data selection/ preprocessing / parameterizations of the LDA model.

0. Data Importing via TDM studio

In [22]:
# Importing our required libraries
import os
import pandas as pd
from lxml import etree
from bs4 import BeautifulSoup
import random
import re 
In [23]:
# Replace the path below with the dataset which you would like to use as input for the script
dataset_directory = 'data/WSJData/'
input_files = os.listdir(dataset_directory)
In [24]:
# Processing large amounts of text can require a lot of memory, and the memory usage of both processing and the resulting 
# dataframe can negatively impact your TDM Studio experience. For this reason, we take a sample of documents
# if there are too many documents in the dataset.
try:
    sample_input_files = random.sample(input_files, 100000)

except ValueError:
    sample_input_files = input_files
In [25]:
## look at all information from one article 
sample_file = dataset_directory + sample_input_files[1]

tree = etree.parse(sample_file)

print('The original xml file \n')
#print(etree.tostring(tree,pretty_print=True))

print('all the elements in the xml \n')
root = tree.getroot()
print([ele.tag for ele in root.iter()])
The original xml file 

all the elements in the xml 

['RECORD', 'GOID', 'Obj', 'SourceRollupType', 'ObjectTypes', 'other', 'mstar', 'ObjectRollupType', 'TitleAtt', 'Title', 'NumericDate', 'StartDate', 'EndDate', 'AlphaDate', 'Language', 'RawLang', 'Language', 'ISO', 'ISOCode', 'ISOExpansion', 'Copyright', 'CopyrightData', 'PrintLocation', 'StartPage', 'Pagination', 'ObjectIDs', 'ObjectID', 'DOCID', 'ObjectID', 'PCID', 'ObjectID', 'PMID', 'ObjectID', 'ProvJournalCode', 'ObjectID', 'PublisherXID', 'Contributors', 'Contributor', 'Author', 'NormalizedDisplayForm', 'LastNameAtt', 'LastName', 'MiddleNameAtt', 'MiddleName', 'FirstNameAtt', 'FirstName', 'OriginalFormAtt', 'OriginalForm', 'Terms', 'GenSubjTerm', 'GenSubjValue', 'Abstract', 'Short', 'AbsText', 'LegacyDataMapping', 'MappingVersion', 'CreatedBy', 'TextInfo', 'Text', 'DFS', 'PubFrosting', 'Title', 'MpubId', 'SortTitle', 'Qualifier', 'Edition', 'JournalCode', 'SourceType', 'StartDate', 'EndDate', 'publisher', 'PublisherName', 'PublisherAddress', 'Address1', 'City', 'Province', 'ZipCode', 'Country', 'URL', 'PublisherNote', 'Locators', 'Locator', 'Name', 'Locator', 'Name', 'CatalogNum', 'Languages', 'Language', 'Flags', 'Flag', 'Flag', 'Flag', 'Flag', 'Flag', 'Flag', 'Flag', 'Flag', 'Flag', 'Flag', 'Flag', 'Subjects', 'Subject', 'CurrentTitle', 'Title', 'SortTitle', 'Qualifier', 'EndIssueDate', 'Edition', 'Locators', 'Locator', 'Name', 'Locator', 'Name', 'CatalogNum', 'CoverImageType', 'BrowseType', 'EmbargoDays', 'HasGaps', 'CoverageRange', 'AlphaStartDate', 'AlphaEndDate', 'NumericStartDate', 'NumericEndDate', 'PubFrequencies', 'ContentModel', 'GroupFrosting', 'AlphaDate', 'StartDate', 'EndDate', 'Locators', 'Locator', 'Name']
In [26]:
# We define a function to get the text content that we need from the XML articles available in our dataset
def getxmlcontent(root):
    if root.find('.//HiddenText') is not None:
        return(root.find('.//HiddenText').text)
    
    elif root.find('.//Text') is not None:
        return(root.find('.//Text').text)
    
    else:
        return None
In [33]:
# Creating three lists to store filename, fulltext, and date
# In TDM studio - the article ID is the same as the filename
filename_list = []
text_list = []
date_list = []
title_list = []
#author_list =[]
subject_list = []
#edition_list=[]
#sourcetype_list = []
#company_list =[]
#test_list = []
subject_dict = []

# Parse files and add data to lists
for file in sample_input_files:
    tree = etree.parse(dataset_directory + file)
    root = tree.getroot()
    
    if getxmlcontent(root) is not None:
        soup = BeautifulSoup(getxmlcontent(root))
        text = soup.get_text()
    else:
        text = 'Error in processing document'
        
    date = root.find('.//NumericDate').text
    #author = root.find('.//Author')
    title = root.find('.//Title').text
    #subject = root.find('.//Subject').text
    #edition = root.find('.//Edition').text
    #source_type = root.find('.//SourceType').text
    #companyname = root.find('.//CompanyName')
    #print(dir(root.find('.//ObjectTypes')))
    
    ## subjects list
    subject_element_list = root.findall('.//GenSubjValue')
    subject_str_list =[] 
    if len(subject_element_list)>=1:
        for i in range(len(subject_element_list)):
            subject_this = subject_element_list[i].text
            subject_str_list.append(subject_this+',')
            subject_dict.append(subject_this)
    else:
        pass
    subject = "".join(subject_str_list)
    
    filename_list.append(file)
    text_list.append(text)
    date_list.append(date)
    title_list.append(title)
    #author_list.append(author)
    #subject_list.append(subject)
    #edition_list.append(edition)
    #sourcetype_list.append(source_type)
    #company_list.append(companyname)
    subject_list.append(subject)
    
    
## get unique subject list
all_subjects = set(subject_dict)
In [34]:
# Creating a dataframe, setting each of the columns to one of the lists we made in the cell above
df = pd.DataFrame({'Article ID': filename_list,
                   'Text': text_list, 
                   'Date': date_list,
                   'Subjects':subject_list,
                   'Title':title_list,
                  #'Edition':edition_list
                  #'SourceType':sourcetype_list,
                  #'CompanyName':company_list
                   #'Test':test_list
                  })
In [35]:
"""
## Here, we use a sub-sample of NYT articles containing key words "inflation" as an example 
## it is saved as a pickle file (a handy and light-weighted data format in python) 
## the path and name

file_name = '../../InfNLPProject/WorkingFolder/Python/article_data.pkl'

article_data = pd.read_pickle(file_name)

"""
Out[35]:
'\n## Here, we use a sub-sample of NYT articles containing key words "inflation" as an example \n## it is saved as a pickle file (a handy and light-weighted data format in python) \n## the path and name\n\nfile_name = \'../../InfNLPProject/WorkingFolder/Python/article_data.pkl\'\n\narticle_data = pd.read_pickle(file_name)\n\n'
In [36]:
## an example of the article 
## the second article in the database and the first 1000 words in the article
## only print part of the article to save the space 
#article_data.iloc[2]['text'][:1000]
df['Text'].iloc[1][:1000]
Out[36]:
'      Belgium\'s Royal Palace raised some eyebrows, and created plenty of chatter on Belgian Sunday TV talk shows, by last week wheeling out 9-year-old Princess Elisabeth for the first formal public appearance of her life, to open a children\'s hospital in her name in Ghent. It ran counter to usual protocol. Under-age Belgian royals are normally only allowed to be seen in public, silently accompanying their parents. So what\'s going on? Pierre-Emmanuel De Bauw, the royal family\'s spokesman, denied any political role for the girl. "The hospital made a request, the parents thought about it and said yes, and the princess accepted," he said. "But it doesn\'t mean she\'ll have a public role in the future." But intentionally or not, a cute 9-year-old making a speech in Dutch is not going to make the royal family less politically popular in Flanders. The royals currently enjoy far less support in Flanders than in French-speaking Wallonia. The most popular Flemish party, the New Flemish Alliance, i'

Start working with the data

In [37]:
article_data = df.copy()
In [38]:
article_data = article_data.rename(columns={'Date':'date',
                                           'Text':'text',
                                           'Title':'title',
                                           'Subjects':'subjects'})

1. Sample Selection

  1. Remove all articles prior to January 1984 and after June 2017.
  2. Exclude articles with page-citation tags corresponding to any sections other than A, B, C, or missing.
  3. Exclude articles corresponding to weekends
  4. Exclude articles with subject tags associated with obviously non-economic content such as sports. List of exclusions available from authors on request.
  5. Exclude articles with the certain headline patterns (such as those associated with data tables or those corresponding to regular sports, leisure, or books columns). (Based on the list provided by authors)
  6. Concatenate articles with the same accession-number as these are chained articles.
  7. Exclude articles with less than 100 words.
In [39]:
article_data
Out[39]:
Article ID text date subjects title
0 1766189757.xml The CNN exit poll for the New Hampshire... 2016-02-19 Political campaigns,Nominations,Presidential e... The Republican Candidates Are Campaigning to Lose
1 888611339.xml Belgium's Royal Palace raised some eyebr... 2011-09-13 Political power, Europe News -- Brussels Beat: Appearance By Pr...
2 1754446997.xml Enjoy traveling easily throughout Europ... 2016-01-08 Refugees,Boundaries,Law enforcement,Sex crimes... Europe's Closing Borders
3 1176675244.xml Your Nov. 13 editorial "America As (Ene... 2012-11-21 Energy industry, Who Cares If the U.S. Beats Arabia?
4 1702330573.xml ATHENS -- Negotiations to secure a thir... 2015-08-10 Bailouts,Eurozone, Europe News: Athens, Creditors Make Progress i...
... ... ... ... ... ...
99995 901622665.xml Syria's government accepted a regionally... 2011-11-03 Violence,Activists,Demonstrations & protests,P... World News: Syria Accepts Arab League Plan
99996 1566536229.xml WASHINGTON -- The White House has moved... 2014-10-01 Intelligence gathering,Presidents,Militancy,In... World News: White House Faces Criticism
99997 1703139871.xml Credit Suisse Group AG is in talks with... 2015-08-12 Dark pool trading,Attorneys general, Credit Suisse in Talks to Settle
99998 1514068776.xml MOSCOW -- The Ukrainian crisis is denti... 2014-04-10 Gross Domestic Product--GDP,Central banks,Sanc... Europe News: Ukraine Crisis Puts Russian Econo...
99999 1220507911.xml Global steel has a big problem: It's to... 2012-11-29 Steel industry,Bankruptcy,Production capacity,... Global Steel Sector Faces Capacity Glut

100000 rows × 5 columns

In [40]:
## set date 
article_data['date'] = pd.to_datetime(article_data['date'],
                                      errors='coerce')
import datetime as dt 
article_data['month_date'] = pd.to_datetime(article_data['date']).dt.to_period('M')
In [41]:
def date_filter(date_st,   ## pd.datetime
                date_ed,
                article_data):
    return article_data[(article_data['date']< date_ed) & (article_data['date']> date_st)] 
In [42]:
def exclude_weekends(article_data):
    article_data['day_of_week'] = article_data['date'].dt.dayofweek
    return article_data[(article_data['day_of_week'] !=5) & (article_data['day_of_week'] !=6)] 
## 5 is saturday and 6 is sunday 
In [43]:
def exclude_subjects(subject,
                     article_data):
    if 'subjects' in article_data.columns:
        return article_data[~article_data['subjects'].str.contains(subject)]
    else:
        return article_data
In [44]:
def exclude_headlines(pattern,
                      article_data):
    if 'title' in article_data.columns:
        return article_data[~article_data['title'].str.lower().str.match(pattern,na=False)]
    else:
        return article_data
In [45]:
def exclude_short(words_limit,
                  article_data):
    article_data['length'] = article_data['text'].str.len()
    return article_data[article_data['length']>=words_limit]
In [46]:
## date filter
date_st = pd.to_datetime('01/01/1984', infer_datetime_format=True)
date_ed = pd.to_datetime('06/01/2017', infer_datetime_format=True)

article_data = date_filter(date_st,
                          date_ed,
                          article_data)
In [47]:
len(article_data)
Out[47]:
95071
In [48]:
## weekends filter 

article_data = exclude_weekends(article_data)
/tmp/ipykernel_24734/3749022856.py:2: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  article_data['day_of_week'] = article_data['date'].dt.dayofweek
In [49]:
len(article_data)
Out[49]:
95027
In [50]:
## filter by length 
article_data = exclude_short(100,
                            article_data)
In [51]:
len(article_data)
Out[51]:
95007
In [52]:
print(all_subjects)
{'Arts festivals', 'Chemical spills', 'Hydrocarbons', 'Development banks', 'Russian literature', 'Francs', 'Metabolites', 'Snowmobiles', 'Building failures', 'Publications', 'Athletic directors', 'Calories', 'Sustainable agriculture', 'Halloween', 'Hairdressers', 'Taiga & tundra', 'Adhesives', 'False information', 'Racetracks', 'Coral reefs', 'Tax sales', 'Hyponatremia', 'Restaurants', 'Foreign students', 'Singers', 'Military strategy', 'Rehearsals', 'Human trafficking', 'Flea markets', 'Bacteria', 'Neighborhoods', 'Physicists', 'Startups', 'Orthodontics', 'Ethnic foods', 'Autophagy', 'Sinkholes', 'International Public Sector Accounting Standards', 'Constitution-US', 'Chemical reactions', 'Cognitive therapy', 'Skiing', 'Tribunals & commissions', 'Architects', 'Silicon wafers', 'Wrongful death', 'LGBTQ students', 'Magnetic resonance imaging', 'Hail', 'Public inquiries', 'Fungicides', 'Smoking', 'Generalized anxiety disorder', 'Saline water', 'Tax controversies', 'Polymer blends', 'Restrictions', 'Strikes', 'Tax rates', 'Pulmonary fibrosis', 'Environmental impact statements', 'Air shows', 'Steroids', 'Lump sum', 'Paperboard', 'Machine learning', 'Magnet schools', 'Sports facilities', 'Willingness to pay', 'Feasibility studies', 'Socialist realism', 'Pesticides', 'Speech', 'Electronic surveillance', 'Blackouts', 'Atherosclerosis', 'State elections', 'Small business', 'Cucumbers', 'Jesus Christ', 'Private schools', 'Cooperatives', 'Family income', 'Economic recovery', 'Personal finance', 'Alligators', 'Financial futures', 'Orphans', 'Unfinished works', 'Stockholders', 'Chaos theory', 'Price stabilization', 'Speculative development', 'Animal shelters', 'Anesthesia', 'Cancer', 'Judicial reviews', 'Market exit', 'Breweries', 'Graphic designers', 'Protective clothing', 'Capital markets', 'Public administration', 'Lubricants & lubrication', 'Success', 'Risk factors', 'Biodegradable materials', 'Customization', 'Renaissance period', 'Border patrol', 'Wealth', 'Potash', 'Seasonal employment', 'Collectibles', 'Aerobics', 'Aviation fuel', 'Candy', 'Trade agreements', 'Coffee', 'Minority students', 'Border security', 'Titanium alloys', 'Investment bankers', 'Tetrahydrocannabinol--THC', 'Receivers & amplifiers', 'Chromosomes', 'Safety programs', 'Cheese', 'Managerial skills', 'Profits', 'Civil society', 'Attorneys', 'Spina bifida', 'Dark matter', 'Luxury homes', 'Billings', 'Medical equipment', 'Professional baseball', 'Tendinitis', 'Nominations', 'VAT', 'Tissue engineering', 'Electronic periodicals', 'Term limitations', 'Credit scoring', 'Multifunctional office equipment', 'Bladder cancer', 'Nurseries', 'Generalized method of moments', 'Special forces', 'Digital photography', 'Statelessness', 'Porous materials', 'Gross income', 'Snack foods', 'Neon', 'Underwater exploration', 'Hinduism', 'Insulin resistance', 'Universalism', 'Caffeine', 'Universal banking', 'Commodities', 'Monitoring systems', 'Juries', 'Tankers', 'Workforce planning', 'Bonuses', 'Brokers', 'Social mobility', 'Corporate image', 'State taxes', 'Tax preparation', '3rd century', 'Pallets', 'Contact lenses', 'Kale', 'Pro choice movement', 'Spiders', 'Family & Medical Leave Act 1993-US', 'Controllers', 'Studies', 'Arms length transactions', 'Power plants', 'Smart cards', 'Medical schools', 'Meat industry', 'Refunds', 'New stock market listings', 'Subways', 'Ecotourism', 'Author productivity', 'Disease transmission', 'Landslides & mudslides', 'Roller hockey', 'Carried interest', 'Urine', 'Age of Enlightenment', 'Databases', 'Radicalism', 'Airport expansion', 'West Nile virus', 'Flavors', 'Government aid', 'Heavy metal music', 'Usability', 'Cod', 'Scholarly publishing', 'Federal court decisions', 'Religious missions', 'Production planning', 'Artificial rubber', 'Descriptive labeling', 'Law enforcement', 'Humidity', 'Lyme disease', 'Psychotherapy', 'Fashion', 'Mindfulness', 'Affordable housing', 'Receipts', 'Caste', 'Silicones', 'Citizenship', 'Use statistics', 'Society', 'Freedom of the press', 'Titanium', 'Shock absorbers', 'Pets', 'Origami', 'Auditions', 'Set design', 'Utopias', 'Austerity policy', 'Bond ratings', 'Jurisprudence', 'Accessories', 'Worms', 'College presidents', 'Forensic accounting', 'Employee complaints', 'Meningitis', 'Minivans', 'British history', 'Net operating losses', 'Remixes (Sound recordings)', 'Beekeeping', 'Junk bonds', 'Company reports', 'Government mandates', 'Interest costs', 'Job sharing', 'Reproductive health', 'Outdoor air quality', 'Catalytic converters', 'Physical therapy', 'Bank holding companies', 'Department stores', 'Accounting systems', 'Small & medium sized enterprises-SME', 'Modems', 'Avant-garde', 'Professionals', 'Telephone companies', 'Earnings', 'Reproductive technologies', 'License plates', 'Long term care insurance', 'Excise taxes', 'Beauty', 'Table tennis', 'Banks', 'Electrocutions', 'Cigarettes', 'Daguerreotypes', 'DNA methylation', 'Cardiovascular disease', 'Funerals', 'Smuggling', 'Geneva Conventions', 'Fees & charges', 'Bathrooms', 'Business to business commerce', 'Brainwashing', 'School closures', 'Pneumonia', 'Biological & chemical weapons', 'Debentures', 'Pilates', 'Fatalities', 'Subconscious', 'Separatism', 'Maternity benefits', 'Multiple listing services', 'School librarians', 'Common Foreign and Security Policy', 'Fair Credit Reporting Act 1970-US', 'Hemorrhoids', 'Pregnancy complications', 'Hospitals', 'Adults', 'Aluminum composites', 'Boilers', 'Customer satisfaction', 'Community development banks', 'National anthems', 'African languages', 'Dental caries', 'Oils & fats', 'Silicosis', 'Deception', 'Telephone hotlines', 'Tinnitus', 'Field programmable gate arrays', 'Foreign exchange controls', 'Correctional personnel', 'Speculation', 'German culture', 'Education savings accounts', 'Compromises', 'Gender reassignment surgery', 'Counties', 'Digital signal processors', 'Gender', 'Emissions', 'Mineral reserves', 'Cyclin-dependent kinases', 'Divisions', 'Geopolitics', 'Artificial sweeteners', 'Procrastination', 'Hackers', 'Shoulder', 'Economics', 'Promotional products', 'Integrated approach', 'Miniatures', 'Mormonism', 'Pesos', 'Group life insurance', 'Religious clothing', 'Dance music', 'Printers', 'Flood insurance', 'Anglophones', 'Deer', 'Gift taxes', 'Nursery rhymes', 'Radio networks', 'Training', 'Meat', 'Perceptions', 'Long term debt', 'Roads & highways', 'Dress codes', 'Servers', 'Statistical analysis', 'Patient admissions', 'Liability insurance', 'Dishonesty', 'Episcopal churches', 'Treason', 'Dishwashing machines', 'Inaugurations', 'Black holes', 'Culture shock', 'Composite materials', 'Satellite dishes', 'Business growth', 'Eye movements', 'Homeless people', 'Victims of crime', 'Drug therapy', 'Upholstery', 'Advertising expenditures', 'Worry', 'Earned income tax credit', 'Viral infections', 'Kings', 'Health care expenditures', 'Aircraft', 'Ride sharing services', 'Bonds', 'Massacres', 'Persian language', 'Water rights', 'Pulmonary hypertension', 'Sports medicine', 'Cannibalism', 'Pipes', 'Pharmacists', 'Sports', 'Serendipity', 'Amusement rides', 'Military prisons', 'Nuclear regulation', 'Tea', 'Chants', 'Sieges', 'Paralympic Games', 'Kidnapping', 'Compensation plans', 'Bands', 'Rare materials', 'Interest income', 'Black athletes', 'Genetic counseling', 'Net income', 'Swimming', 'Probiotics', 'Contamination', 'Martial law', 'Vodka', 'Canals', 'Diplomacy', 'Off-Broadway theater', 'Limited partnerships', 'Surgeons', 'Cystic fibrosis', 'Evolution', 'Business travel', 'Merchant banks', 'Remittances', 'Public officials', 'Consortia', 'Keynesian theory', 'Generational differences', 'Logistics', 'Contemporary problems', 'Loan agreements', 'Demonstrations & protests', 'Immunoassay', 'Stereotypes', 'Islamic studies', 'Mammals', 'Nutrition', 'Installation art', 'Warranties', 'Hypothermia', 'Arthritis', 'Music', 'Petroleum production', 'Golden parachutes', 'Air fares', 'Moving & housing expenses', 'Political alliances', 'Botulinum toxin', 'Histamine', 'Ballroom dancing', 'Income redistribution', 'Medical ethics', 'Florists', 'Overweight', 'Ornithology', 'Transnationalism', 'Drug prevention', 'Insurgency', 'School boards', 'Mobility', 'Neutrality', 'Dynasties', 'Labeling', 'Sunscreen', 'School principals', 'First families', 'Liberalism', 'Phonetics', 'Miscarriage', 'Travel insurance', 'Hate crimes', 'Bacterial infections', 'National scenic areas', 'Rebates', 'Legal aid', 'Congressional powers', 'Ecstasy', 'Bank deposits', 'Planetariums', 'Political history', 'Evictions', 'Electronic filing', 'Diet', 'Habitats', 'Corporate banking', 'Self esteem', 'Collage', 'Repair & maintenance', 'Ethanol', 'Same sex marriage', 'Working capital', 'Keratin', 'Snowshoes', 'Fluorides', 'Cement', 'Muslims', 'Quantitative genetics', 'Black nationalism', 'Hairstyles', 'Stigma', 'Structured settlements', 'Zionism', 'Media rights', 'Pneumothorax', 'Perjury', 'Network security', 'Enforcement', 'Falls', 'Golf equipment', 'Civil judgments', 'Rubella', 'Joint and ligament injuries', 'Energy tax credit', 'Consumer-driven health plans', 'Bass instruments', 'Pulp & paper mills', 'Immunology', 'Tidal waves', 'Cosmetics industry', 'Bone surgery', 'Caves', 'Guarantees', 'Pressure vessels', 'Kyoto Protocol', 'Marital separation', 'Research funding', 'Nonalcoholic beverages', 'Nonferrous metals', 'Debit cards', 'Nicotine', 'Leather & leather products', 'Accounts receivable', 'Accounting firms', 'Tagging', 'Civil service', 'Womenswear', 'Oil consumption', 'Wealth management', 'Business indicators', 'Plant reproduction', 'Motion capture', 'Eggs', 'Military sales', 'Reference books', 'Reunification', 'Outdoor activities', 'Antimicrobial agents', 'Bulletin boards', 'Multinational space ventures', 'Crime drama', 'Adenoviruses', 'Choirs', 'Debt financing', 'Native American studies', 'Thermal energy', 'Indigent care', 'Coalition governments', 'Orienteering', 'Extractive industries', 'Stem cells', 'Veins & arteries', 'Special drawing rights', 'Public prosecutors', 'Voice recognition', 'Ballooning', 'Ovulation', 'Holiday decorations', 'Varicose veins', 'Kerosene', 'Creative industries', 'Vibration', 'Reincarnation', 'Dairy cattle', 'Misogyny', 'Mass murders', 'Food products', 'Magazine industry', 'Desserts', 'Public art', 'Adultery', 'Soft skills', 'Time series', 'Foreign Corrupt Practices Act 1977-US', 'Operating costs', 'Illustrations', 'Advanced manufacturing technologies', 'Back surgery', 'SEC registration', 'Corruption in government', 'Dermatology', 'Recording sessions', 'Espionage', 'Software upgrading', 'Authoritarianism', 'Cultural appropriation', 'Powers of attorney', 'Trends', 'Work ethic', 'Etymology', 'Related party transactions', 'Cost control', 'Body piercing', 'JOBS Act-2012', 'Dementia', 'Fixed exchange rates', 'Job titles', 'Community colleges', 'Licensing', 'Electronic games', 'Medical records', 'Alcohol', 'Publishing industry', 'Public access', 'Commercial aircraft', 'Court hearings & proceedings', 'Counseling services', 'State budgets', 'Polyethylene terephthalate', 'School nurses', 'Agriculture', 'Grass roots movement', 'CRISPR', 'Pension funds', 'Wheat', 'College graduates', 'Deregulation', 'Fish hatcheries', 'Maritime industry', 'Children & youth', 'Safety standards', 'Dilution', 'Genetic engineering', 'Musicians & conductors', 'Civil rights', 'Powertrain', 'Jargon', 'Centuries', 'Gas detectors', 'Transfer of funds', 'Resource recovery', 'Free trade', 'Mathematics teachers', 'Arms control & disarmament', 'Cardiac arrhythmia', 'Woodcuts', 'Refugees', 'Video industry', 'Ownership changes', 'Incineration', 'Emergencies', 'Immigration', 'Terrorism financing', 'Deposit accounts', 'Fractional aircraft ownership', 'Acquisitions & mergers', 'Stagnation', 'Tourist attractions', 'Sharks', 'Insurance premiums', 'Steel', 'Demutualization', 'Underfunded pension plans', 'Covenants', 'Digital electronics', 'Length of stay', 'Legal medicine', 'No confidence motions & votes', 'Correspondence', 'Toilet facilities', 'Privileges & immunities', 'Web sites', 'Sex industry', 'Liberation theology', 'Inventory', 'Immunization', 'Filtering systems', 'Insurance companies', 'Carbon monoxide', 'Lifetime', 'Individualism', 'Fever', 'Closing the sale', 'Ink jet printers', 'Prime ministers', 'Trains', 'Defections', 'Candles', 'Bank liquidity', 'Automobile customizing', 'Keyboards', 'Sound', 'Hearing protection', 'Digital cameras', 'College hockey', 'Solvents', 'General anesthesia', 'Soy products', 'Fiber optic networks', 'Playgrounds', 'French Revolution', 'Rap musicians', 'Boundaries', 'Information overload', 'Tires', 'Cruelty to animals', 'Imperialism', 'Blood vessels', 'Nomads', 'Home ownership', 'Landscape art', 'Fertility', 'Free enterprise', 'Penis', 'Radio programming', 'Construction spending', 'Export credit insurance', 'Electronic commerce', 'Sport clothes', 'Biohazards', 'State police', 'Electricity', 'Police stations', 'Business ethics', 'Cubism', 'Professional golf', 'Complaints', 'C corporations', 'Market entry', 'Zoning ordinances', 'Demographics', 'Retirement benefits', 'Stimulants', 'Cuban Missile Crisis', 'Curricula', 'Hand surgery', 'Waterways', 'Subcontractors', 'Assimilation', 'Consumer credit', 'Good samaritans', 'Disinfection & disinfectants', 'Patient compliance', 'Fault lines', 'Political appointments', 'Water shortages', 'Exchange programs', 'Media planning & buying', 'Diversion programs', 'Gossip', 'Biomedical research', 'Special purpose entities', 'Corporate growth', 'Hygiene', 'Mapping', 'Rosh Hashanah', 'Salt', 'Aquifers', 'Portfolio management', 'Female employees', 'Gin', 'Migrant workers', 'Activists', 'National debt', 'Aluminum', 'Natural & organic foods', 'Sign language', 'Disaster recovery', 'Nuclear fuels', 'Credit card processing', 'Garage sales', 'Economic aid', 'Proprietary', 'Censuses', 'Chambers of commerce', 'Vertical integration', 'Service introduction', 'Yogurt', 'National conservation areas', 'Butter', 'Nonprofit hospitals', 'Costumes', 'Construction contracts', 'Loan losses', 'Academy awards', 'Water conservation', 'Interstate commerce', 'Abscesses', 'Femininity', 'Religious exemptions', 'Burden of proof', 'Adjustable rate mortgages--ARM', 'Dissection', 'Quality of life', 'Stakeholders', 'Group therapy', 'Jewish law', 'Indoctrination', 'Multiculturalism & pluralism', 'Climate change', 'Medicare Prescription Drug Improvement & Modernization Act 2003-US', 'Product testing', 'Arab Spring', 'Elephants', 'Graphene', 'Mediators', 'Ancient civilizations', 'Foraging behavior', 'Motion picture directors & producers', 'Womens rights movement', 'Terrorism', 'Stream water', 'Dietary restrictions', 'Summer solstice', 'Information society', 'District courts', 'Word of mouth advertising', 'Linguistics', 'Civil disobedience', '11th century', 'Branch banking', 'Domestic service', 'Memorial services', 'Business models', 'Environmental justice', 'Squatters', 'Default', 'Career changes', 'Waterfront development', 'Peasants', 'Neocolonialism', 'Land purchases', 'Information technology', 'Animal protection', 'Kitsch', 'Independence Day', 'Police pursuit driving', 'Relationship banking', 'Soprano voice', 'Value chain', 'Insurance agents & brokers', 'Football', 'Radishes', 'Prime rate', 'Respiration', 'Expansion', 'Health & beauty aids', 'Tariffs', 'Short term debt', 'Acupuncture', 'Accounting records', 'Extremism', 'Repertoire', 'Resignations', 'Hair', 'Chinese languages', 'Women', 'Severance taxes', 'Informatics', 'Disability discrimination', 'Free agency', 'Paleontology', 'Luge', 'Fugitives', 'Militarism', 'S corporations', 'Search & seizure', 'Prostate cancer', 'Presidential communications and messages', 'Compensation', 'Hunter-gatherers', 'Military engagements', 'Film schools', 'Air traffic control', '19th century', 'Coalbed methane', 'University administration', 'Metadata', 'Time travel', 'Stem cell transplantation', 'Docks', 'Safety research', 'Response rates', 'Dental floss', 'Fibroids', 'Physical education teachers', 'Preeclampsia', 'Coding standards', 'Atmosphere', 'Life', 'Hybrid vehicles', 'Esports', 'Bedouins', 'Paganism & animism', 'Climate', 'Federal aid', 'Graphic arts', 'Office layout', 'Crop insurance', 'Exchange traded funds', 'Dyskinesia', 'Animal reproduction', 'Demolition', 'Ryan White Comprehensive AIDS Resources Emergency Act 1990-US', 'Starvation', 'Job classification', 'Television advertising', 'Karaoke', 'Mineral resources', 'Comics', 'Business improvement districts', 'Teleconferencing', 'Sugar industry', 'Aneurysms', 'Popes', 'Expiration', 'Airships', 'Birds', 'Pancreas', 'Bank officers', 'Soap operas', 'Sting operations', 'Nails (Anatomy)', 'War crimes', 'Human-computer interaction', 'Polychlorinated biphenyls--PCB', 'Community policing', 'Republicanism', 'Talent management', 'Mortgage brokers', 'Conventions', 'Flat panel displays', 'Toll roads', 'Population', 'Cosmology', 'Couples', 'Negative campaigning', 'Old English', 'Volatile organic compounds--VOCs', 'Pipelines', 'Esophagus', 'Candy industry', 'Life insurance', 'Language arts', 'Explorers', 'Disease control', 'Food chains', 'Rural development', 'OEM', 'Auditors', 'Bronze', 'Alkalinity', 'Presbyopia', 'Corporate profits', 'Sewage disposal', 'Crop dusting', 'Sleeping bags', 'Geriatrics', 'Anger', 'Hogs', 'Food programs', 'Advertising agencies', 'Biological rhythms', 'Ukulele music', 'Damage claims', 'Biogas', 'Dentists', 'Carrots', 'Case studies', 'Geographic information systems', 'Wild and scenic rivers', 'Cost analysis', 'Antisocial personality disorder', 'Iron', 'Crimes against public order', 'Adult education', 'Cloning', 'Strategic materials', 'Nontraditional students', 'Retention', 'Individual retirement accounts--IRA', 'Detention centers', 'Insurance syndicates', 'Black markets', 'Public utilities', 'Consumerism', 'Disposable income', 'Pepper spray', 'Robbery', 'Physiology', 'Disabled workers', 'Self destructive behavior', 'Vaudeville', 'Planning', 'Goats', 'Gag orders', 'Water heaters', 'Liver cancer', 'Sales', 'Prison escapes', 'Brands', 'Vibrato', 'Datasets', 'Heart failure', 'Freight forwarding', 'Smallpox', 'Hotels & motels', 'Food contamination & poisoning', 'Careers', 'Inspections', 'Values', 'Biomechanics', 'Bodyguards', 'Presentations', 'Trademarks', 'Recall', 'Furnishings', 'Electronic discovery', 'Sailors', 'Comic books', 'Gamma rays', 'Third party administrators', 'Group homes', 'Morality', 'Industrial design', 'Mineralogy', 'Painting', 'Credit card fraud', 'Alternative sentencing', 'Budget Control Act 2011-US', 'Digital signatures', 'Insulin-like growth factors', 'Amphetamines', 'Energy policy', 'Trauma centers', 'Legal documents', 'Food security', 'Labor productivity', 'Rainbows', 'Achievement tests', 'Working mothers', 'Electric industries', 'Windfall profits', 'Marriage', 'Rocket launches', 'Resorts & spas', 'Prepayments', 'Airline security', 'Gorillas', 'Convulsions & seizures', 'Latin American literature', 'Compasses', 'Loss recognition', 'Subscriptions', 'Alto voice', 'Family planning', 'Bilingualism', 'Stormwater', 'Current liabilities', 'Energy exports & imports', 'Neuroses', 'Business plans', 'Humanities', 'Hominids', 'Falklands War', 'Flight training', 'Midwifery', 'Pluto', 'Experimental psychology', 'Activism', 'Online instruction', 'Labor law', 'Brand image', 'Costs', 'Electronics industry', 'Vigilantes', 'Commercial credit', 'Language instruction', 'Imports', 'Stained glass', 'Correctional treatment programs', 'Dance festivals', 'Mastectomy', 'Anti-communism', 'Denial of service attacks', 'Cell division', 'Disabled students', 'Space weapons', 'Bids', 'Ethernet', '20th century', 'Vacations', 'Source studies', 'Gastroesophageal reflux', 'Mines', 'Immigrant students', 'Leasing companies', 'Traumatic brain injury', 'Fish', 'Orgasm', 'Stone', 'Risk exposure', 'Grammy awards', 'Audio recordings', 'Caskets', 'Infrastructure', 'Adaptations', 'Incentives', 'Baptism', 'Child marriage', 'Dams', 'Limit of liability', 'Prenatal development', 'Foxes', 'Trouble shooting', 'Gas turbine engines', 'Electrocardiography', 'Value creation', 'Postal codes', 'Queuing theory', 'Alternative energy sources', 'Petrochemicals', 'Centenarians', 'Anglican churches', 'Waxes', 'Mobile businesses', 'Lean manufacturing', 'Language acquisition', 'College admissions', 'Crude oil', 'Environmentalists', 'Patients', 'Hobbies', 'Domestic partners', 'Tony awards', 'Psychosomatic medicine', 'Developmental disabilities', 'Military air strikes', 'Prostitution', 'Holy places', 'Moose', 'Endangered & extinct species', 'Terminal illnesses', 'Information communication', 'Tutoring', 'Appendicitis', 'High school basketball', 'Motorcycles', 'Workshops', 'Childrens literature', 'Disabled people', 'Right to know', 'Condoms', 'Public domain', 'Apoptosis', 'Francophones', 'Skill development', 'Bankruptcy', 'End users', 'Long term', 'Scholarly communication', 'Background checks', 'Remakes & sequels', 'Intellectual disabilities', 'Dyslexia', 'Silent films', 'Tin', 'Mathematicians', 'Antitrust laws', 'Military housing', 'Eveningwear', 'Pragmatism', 'Cracks', 'Casting directors', 'Telecommunications policy', 'Personal protective equipment', 'Proxy solicitation', 'Toddlers', 'Automobile drivers', 'Timberlands', 'Radiation', 'Cities', 'Constitutional law', 'Loopholes', 'Epic literature', 'Personal bankruptcy', 'Stock splits', 'Afrikaans language', 'Fiction', 'Drying agents', 'Ceramics', 'Cultural differences', 'Adaptive immunity', 'Tax shelters', 'Amendments', 'Fireworks', 'Rowing', 'Arrest warrants', 'Cost of living', 'Programmers', 'Sparkling wine', 'Oil reserves', 'Favoritism', 'Artificial satellites', 'Self sufficiency', 'Behavior modification', 'Messianism', 'Presidential protection', 'Telescopes', 'RICO 1970-US', 'Police corruption', 'Defense', 'Running', 'Landowners', 'Campuses', 'Anti inflation', 'Childrens picture books', 'Listening', 'Earthmoving equipment', 'Bone marrow', 'Berries', 'Firearm discharge residue', 'Emancipation of slaves', 'Premature birth', 'English teachers', 'Track & field', 'Speed skating', 'Weddings', 'Quartz', 'Vitamin E', 'Greeting card industry', 'Quantum physics', 'Sexes', 'Air traffic controllers', 'Questioning', 'Broadcasting industry', 'Electroconvulsive therapy', 'Sexual intercourse', 'Optimism', 'Time zones', 'Chief technology officers', 'Printed circuit boards', 'Community theater', 'Cheerleaders', 'Funk music', 'Industrial production', 'Equal rights', 'Air travel', 'Commercial fishing', 'Drug dosages', 'Seismic engineering', 'Sports marketing', 'Grand juries', 'Construction', 'Anti-virus software', 'Lighting systems', 'Experiments', 'Denim', 'Schedules', 'Commodity prices', 'Tunnels', 'Fair market value', 'Abstract art', 'Preventive medicine', 'Pandemics', 'Overspending', 'Breast implants', 'Artificial turf', 'Stormwater management', 'Military pay', 'Scientific method', 'Intelligence', 'Credit card industry', 'Fisheries management', 'Remote control', 'Internet stocks', 'Genital mutilation', 'Mammography', 'Speakers of the House', 'Searches', 'Space shuttle', 'Natural gas reserves', 'Riverboat casinos', 'Inuit', 'Creativity', 'Peer to peer computing', 'Earth', 'Asians', 'Natural gas distribution', 'Financial accounting standards', 'Prehistoric era', 'Middle East respiratory syndrome', 'Insect control', 'Editorial cartoons', 'Firearm laws & regulations', 'Subprime lending', 'Investment policy', 'Home births', 'Ears & hearing', 'Water treatment', 'Academic freedom', 'Antebellum period', 'Dividend distributions', 'Internet of Things', 'Reactors', 'Sunglasses', 'Neurology', 'Economic summit conferences', 'Property damage', 'Windows operating system', 'Sinusitis', 'Annual meetings', 'El Nino', 'Interest rate swaps', 'Chemistry', 'Anarchism', 'Educational technology', 'Science', 'Big Four accounting firms', 'Vases', 'Legs', 'Auditors reports', 'Households', 'Autonomous vehicles', 'Automobile safety', 'Popularity', 'Philology', 'Back pain', '9th century', 'Government shutdowns', 'Revocable trusts', 'Distance learning', 'Cholera', 'Annealing', 'Animal care', 'Deconstruction', 'Witchcraft', 'Talent agents & managers', 'Social investing', 'Turnaround management', 'Lumber industry', 'Erectile dysfunction', 'False alarms', 'Sleep disorders', 'Headphones', 'Ophthalmology', 'Purchasing managers index', 'Site planning', 'Songwriters', 'Sexual orientation', 'Wholesale clubs', 'Biotechnology', 'Participating loans', 'Theological schools', 'Appropriations', 'Cartoons', 'Racial discrimination', 'Muslim Americans', 'Small kitchen appliances', 'Colonialism', 'Cadavers', 'Trusts', 'Deserts', 'Rhinitis', 'Web portals', 'Workplace diversity', 'Vegetable juices', 'Simulation', 'Racial justice', 'Cost accounting', 'Electric vehicles', 'Powers of appointment', 'Anemia', 'Discretionary power', 'Savings plans', 'Saturn', 'Teenagers', 'Bank robberies', 'Vitamin C', 'Classified advertising', 'Student debt', 'Fines & penalties', 'Office supplies', 'Textile fibers', 'Bailouts', 'Debt', 'Disk drives', 'Licenses', 'Drug withdrawal', 'Trust departments', 'Armenian Americans', 'Ticket sales', 'Armored vehicles', 'Correctional institutions', 'Food science', 'Radar', 'Nickel', 'Discount department stores', 'Turbines', 'Mental competency', 'Monetary unions', 'Wineries & vineyards', 'Racial identity', 'Formal organization', 'Future of banking', 'Coma', 'Skilled workers', 'Oil & gas royalties', 'Service industries', 'Federal courts', 'Medals', 'Digital transmission', 'Electric currents', 'Volcanoes', 'Rare books', 'Bubbles', 'Glucose', 'Theater revivals', 'Oysters', 'Phosphates', 'Nudes', 'Web site design', 'Typhoid', 'Military exercises', 'Factory farming', 'Multiracial people', 'Opposition parties', 'Fossil fuels', 'Construction equipment', 'Revolving credit', 'Mens health', 'Taxation economics', 'Internet', 'Foreign business', 'Foreign subsidiaries', 'Zombies', 'Morphine', 'Shakeups', 'Marsupials', 'Trespassing', 'Disclosure', 'Food quality', 'Universal Serial Bus', 'Manufactured products', 'Persian Gulf War', 'Colon', 'Yarn', 'Cybernetics', 'Protestantism', 'Capital punishment', 'Strip mining', 'Muscle pain', 'Monopolistic competition', 'Environmental law', 'Catholicism', 'String theory', 'Cartels', 'Payroll taxes', 'Sociology', 'Sperm', 'Smoke inhalation', 'Traffic', 'Cardiac function', 'Product reviews', 'Land reform', 'Artisans', 'Taxation', 'Coffeehouses', 'Social security taxes', 'Vitamin B', 'Private banking', 'Minority owned businesses', 'Provisions', 'Wireless networks', 'Leaves', 'Love', 'Myasthenia gravis', 'Americans with Disabilities Act 1990-US', 'Militia groups', 'Reductionism', 'College sports', 'Deferred compensation', 'Bombs', 'Financial planners', 'Executive function', 'Pessimism', 'Political risk', 'Sunday legislation', 'Knots', 'Cattle industry', 'Anti-Semitism', 'Calcium phosphates', 'Research parks', 'Headaches', 'Metaphysics', 'Print on demand', '3-D graphics', 'Managers', 'Trapping', 'Airlifts', 'Weeds', 'Play therapy', 'Respiratory diseases', 'Therapy', 'Automotive supplies', 'Particle accelerators', 'Buffalo', 'Home education', 'Repatriation', 'Communications networks', 'Cytomegalovirus', 'Architecture', 'School libraries', 'Wound healing', 'Dietary minerals', 'Lysergic acid diethylamide--LSD', 'Hazing', 'Minority & ethnic groups', 'Biopsy', 'Electronic cigarettes', 'Theaters & cinemas', 'Negotiations', 'Atheism', 'Alcohol use', 'Customer relationship management', 'Interviews', 'Wage rates', 'Aeronautics', 'Neural networks', 'Kurds', 'Exclusive economic zone', 'Astronauts', 'Cancer therapies', 'Biology', 'Deportation', 'Urology', 'Credit ratings', 'Generic products', 'Yield', 'Engineering schools', 'Parachuting', 'Elitism', 'Comedies', 'Automobile loans', 'Convictions', 'Mountains', 'Adult children', 'Reputation management', 'Government employees', 'Interest rates', 'Salary caps', 'Police brutality', 'Preferred stock', 'Physicians', 'Garages', 'Sleep', 'Avocados', 'Fire protection', 'Regions', 'Liability', 'Security services', 'Segregation', 'Carbon', 'Food trucks', 'Geography', 'Ecologists', 'Carnivals', 'Rites & ceremonies', 'Financing leases', 'Molecular biology', 'Lead poisoning', 'Stock offerings', 'Atmospheric pressure', 'Trade liberalization', 'Public opinion surveys', 'Musical theater', 'Retirement plans', 'Animal feathers', 'Taxicabs', 'Recipes', 'Fashion designers', 'Hydrotherapy', 'Child soldiers', 'Cash registers', 'Private sector', 'Colleges & universities', 'Aviation', 'Public transportation', 'Partnerships', 'Pre-existing conditions', 'Employment agencies', 'Liver diseases', 'State court decisions', 'Novellas', 'Cervix', 'Crime scenes', 'Forestry', 'Electricians', 'Universal service', 'Urination', 'Test systems', 'Invasion of privacy', 'Lost & found property', 'Clean Water Act-US', 'Nationalism', 'Microbreweries', 'Natural gas vehicles', 'Report writing', 'Arson', 'Vision systems', 'Mardi Gras', 'Pedophilia', 'Wind', 'Mortgage servicing', 'Qualifications', 'Childrens health insurance programs', 'Public safety', 'Sunset provisions', 'Bad debts', 'Nielsen ratings', 'Sport science', 'Virtual offices', 'Footwear industry', 'National libraries', 'Down syndrome', 'Health clubs', 'Excommunication', 'Environmental protection', 'Zika virus', 'Sports officiating', 'Defense contracts', 'Anaphylaxis', 'Prevention', 'Cataracts', 'Deferred income', 'Diesel fuels', 'Disaster insurance', 'Balloons', 'Strawberries', 'Government waste', 'Legalized gambling', 'Network operating systems', 'Tap dance', 'Nitrates', 'Growth rate', 'Cosmetics', 'Immune system', 'Turkish culture', 'Pattern recognition', 'Quality control', 'Leadership', 'Nutritionists', 'Gene therapy', 'Melatonin', 'Internet resources', 'Nazi era', 'Bond markets', 'Sanctuary movement', 'Illnesses', 'Automotive parts industry', 'Geodetics', 'Compatible hardware', 'Retina', 'Religious wars', 'Computer centers', 'Guaranty funds', 'Chiropractic medicine', 'Workforce', 'Production capacity', 'Dust', 'Industrial espionage', 'Snowboarding', 'Field study', 'Gynecology', 'Drug addiction', 'Biomarkers', 'Oilseeds', 'Insurance agencies', 'Halal food', 'Fraud', 'Control theory', 'Foreclosure', 'Oxidation', 'Multiple dwellings', 'Embassy security', 'Court decisions', 'Rating services', 'Storytelling', 'Cooling', 'Inner city', 'Saints', 'Womens history', 'Criminal pleas', 'Prefabricated buildings', 'Menswear', 'Specialty services', 'Alumina', 'Onions', 'Missiles', 'Defense spending', 'Vocal music', 'Dissertations & theses', 'Euroscepticism', 'Education expenses', 'Plate tectonics', 'Desegregation', 'Discount rates', 'Oil pollution', 'Mezzanine financing', 'Ordination', 'Human nature', 'Community centers', 'Loss reserves', 'Tax collections', 'Economic models', 'Plywood', 'Departments', 'Relocation', 'Pretrial discovery', 'Prospecti', 'Solar energy', 'Carotid arteries', 'Chargebacks', 'Biometric identification', 'Emergency communications systems', 'Tranquilizers', 'Agricultural economics', 'Building materials industry', 'Text analysis', 'Toothpaste', 'Naval vessels', 'Large cap investments', 'Phenols', 'Ancient languages', 'Bank fraud', 'Horses', 'Internal medicine', 'Linux', 'Business education', 'Reporting requirements', 'Telephone directories', 'Automobile rentals', 'Orientations', 'Accuracy', 'Diuretics', 'Credit cards', 'International cooperation', 'Five year plans', 'Search engines', 'Botulism', 'Accreditation', 'Embryos', 'Printing industry', 'Private enterprise', 'Used goods', 'Nostalgia', 'New year', 'Advocacy', 'Public private partnerships', 'Food stamps', 'Network switching', 'Ice ages', 'Trucks', 'Mothers', 'Entrapment', 'Early childhood education', 'Direct marketing', 'Autumn', 'Kiosks', 'Trans fats', 'Smoke detectors', 'Cable TV', 'Treaties', 'Ice hockey', 'Telematics', 'Ballet', 'Cash management', 'Labor relations', 'Cruise lines', 'Ex-convicts', 'Domestic markets', 'Islamic art', 'Blu-ray discs', 'Information industry', 'Rands', 'Trumpet music', 'Chairman of the board', 'Monks', 'Baptist churches', 'Rationalism', 'Mail fraud', 'Religious congregations', 'Entomology', 'Operating leases', 'Supply & demand', 'New Testament', 'Sawmills', 'Sports drinks', 'Riot control', 'Shopping', 'Joint ventures', 'Anthrax', 'Indecent exposure', 'Public sector', 'Population growth', 'Economic development', 'Librarians', 'Rock climbing', 'Defensive stocks', 'Juvenile courts', 'Stock prices', 'Market shares', 'Vaccines', 'Labor market', 'Prison industries', 'Tactical units', 'Monetary policy', 'Prepaid debit cards', 'Computer networks', 'Murals', 'Homemakers', 'Democracy', 'Ground leases', 'Hypoxia', 'Biathlon', 'Interior design', 'Reservoirs', 'Timber', 'Animal rights movement', 'Routers', 'Guillain-Barre syndrome', 'New Deal', 'Self defense', 'Reparations', 'Postal & delivery services', 'Personal computers', 'Property management', 'Wire', 'Crop damage', 'Accident prevention', 'Skin', 'Neuroendocrine tumors', 'Soil fertility', 'Developers', 'Diversity training', 'Clientelism', 'Land surveys', 'Biological products', 'Koran', 'Outdoor furniture', 'Product mixes', 'Flashlights', 'Remedial education', 'Free markets', 'College campuses', 'Trade restrictions', 'Renovation & restoration', 'Water mains', 'Classified information', 'Superannuation', 'Special elections', 'Receivership', 'Postcolonialism', 'Antiques', 'British & Irish literature', 'Specialty products', 'International education', 'Breakdowns', 'Ontology', 'Regional Comprehensive Economic Partnership', 'Paints', 'Skin care products', 'Expatriates', 'Hysterectomy', 'Schools', 'Petroleum refineries', 'Homework', 'Licensed products', 'Zero coupon bonds', 'Debt management', 'Voice response technology', 'Labor force', 'Regulation D', 'Bad faith', 'Luxury taxes', 'Tax liens', 'Classrooms', 'Olympic trials', 'Metric system', 'Screen time', 'Trainers', 'Member services', 'Judges & magistrates', 'Load', 'Westernization', 'Teacher evaluations', 'Ticket scalping', 'Impaired assets', 'Building permits', 'Data encryption', 'Burlesque', 'Corporate credit cards', 'Eminent domain', 'Qualified tuition programs', 'Hunger strikes', 'Librettists', 'Identification systems', 'Job hunting', 'TARP funds', 'Federal Housing Enterprises Financial Safety & Soundness Act 1992-US', 'Reporters', 'Parliamentary procedure', 'Dietary fiber', 'Welfare', 'Hunting', 'Blister packs', 'Transparency', 'Drilling', 'Hospital systems', 'Nursing', 'Cholesterol', 'Water measurement', 'Social anxiety', 'Freelance', 'Earnings trends', 'Information services', 'Survival analysis', 'Lunar eclipses', 'Serial crime', 'Federal employees', 'Restitution', 'Kites', 'Type fonts', 'Brand management', 'Reproductive rights', 'Rohingya', 'Pregnancy', 'Vacuum cleaners', 'Creative process', 'Operating ratios', 'Private libraries', 'Democratization', 'Automobile shows', 'Estate taxes', 'Voter behavior', 'Trading cards', 'Coping', 'Smooth muscle', 'Target markets', 'Quotas', 'Wholesale Price Index-US', 'Chief executive officers', 'Fish oils', 'Monkeys & apes', 'Brain research', 'In vitro fertilization', 'Islam', 'Pastures', 'Committees', 'Entrance examinations', 'Right to die', 'Pictographs', 'Plant-based foods', 'Legislators', 'Hardware reviews', 'Belly dancing', 'Web archiving', 'Polygraphs', 'Lymphocytes', 'Governor General-Canada', 'Sedition', 'Medical device industry', 'Intellectual freedom', 'Commercials', 'Commodities trading', 'Rationality', 'Physical sciences', 'Art Nouveau', 'Stadiums', 'Ferries', 'Petroleum engineering', 'Property & casualty insurance', 'Cooperative learning', 'Pharmacy', 'Bureaucrats', 'Autocracy', 'Computer viruses', 'Infringement', 'Social isolation', 'Problem solving', 'Surround sound', 'Females', 'Travel agencies', 'Social security', 'Sarcoma', 'Floods', 'Breastfeeding & lactation', 'Tents', 'Building societies', 'USA PATRIOT Act 2001-US', 'Nursing care', 'Commuting', 'Trios (Performing ensembles)', 'Television news', 'Research centers', 'Environmental conditions', 'Residential buildings', 'Computer & video games', 'International law', 'Laboratories', 'Fiscal policy', 'Kidney transplants', 'Master limited partnerships', 'Whiskey', 'Young adults', 'Political finance', 'Sovereign wealth funds', 'Raspberries', 'Patient Protection & Affordable Care Act 2010-US', 'Hurricanes', 'Direct democracy', 'Foreign tax credits', 'Audit trails', 'Abolition of slavery', 'Gaming machines', 'Missile defense', 'Corruption', 'Police departments', 'Interpreters', 'Family medical history', 'Ship seizures', 'Graphite', 'Taste', 'Statutory rape', 'Recusal', 'Biodiversity', 'Urbanism', 'Priorities', 'Independent contractors', 'Military reserves', 'Disasters', 'Hit & run accidents', 'Airlines', 'Bees', 'Trucking', 'Natural history', 'Age', 'State courts', 'Monte Carlo simulation', 'Vital signs', 'Fruits', 'Stomach cancer', 'Aircraft accidents & safety', 'Hypnosis', 'Apartments', 'Information warfare', 'Mineral water', 'Export taxes', 'Jewish Americans', 'Storm damage', 'Urban areas', 'Womens health', 'Subcultures', 'Photofinishing laboratories', 'Cellulose acetate', 'Product placement', 'Localization', 'Private property', 'Insured losses', 'Multiple myeloma', 'General Data Protection Regulation', 'Economic forecasts', 'Voir dire', 'Aesthetics', 'Video art', 'Posture', 'Displaced workers', 'Fraternities & sororities', 'Human relations', 'Consolidation', 'Great Recession', 'Frequent flier programs', 'Vegetation', 'Drug crimes', 'Nanowires', 'Theologians', 'Artists-in-residence', 'University graduates', 'Postal rates', 'HD-DVD', 'Facilities planning', 'Alzheimers disease', 'Real estate financing', 'Wildfowl', 'Vertebra', 'Mail order', 'Nanotechnology', 'Growth funds', 'Mascots', 'Food', 'Ocean waves', 'Loans', 'Rebellions', 'Visual artists', 'Intimacy', 'Conversion', 'Semiconductors', 'Credit management', 'Chronic illnesses', 'Human error', '4th century', 'Counterterrorism', 'Wellness programs', 'Neurosciences', 'Apportionment', 'Academic standards', 'Breach of contract', 'Cross country skiing', 'Art theft', 'Motorcycle racing', 'Fractional interests', 'Rotator cuff', 'Swimwear', 'Audit committees', 'Prostate', 'Biologists', 'Border walls', 'Defined benefit plans', 'Resale value', 'Pancreatitis', 'Animal training', 'International organizations', 'Tax incentives', 'Audiences', 'Typhus', 'Consumer Price Index', 'Iron compounds', 'Shortages', 'Uterus', 'Silver mines', 'Skin diseases', 'Actors', 'Black owned businesses', 'Political leadership', 'Farming', 'Telemedicine', 'Medicaid', 'Laser surgery', 'Freshwater resources', 'Recreation', 'Natural law', 'Post offices', 'Dialects', 'Geometry', 'Topography', 'Poliomyelitis', 'Transponders', 'Government documents', 'Communion', 'Hydrocephalus', 'Manufacturing', 'Extradition', 'Gender dysphoria', 'Raw materials', 'Viruses', 'Theater directors & producers', 'White supremacists', 'Elementary school teachers', 'Alimony', 'Open source software', 'Computer privacy', 'Tax assessments', 'Extortion', 'Supervision', 'Scheduling', 'Student participation', 'Expected values', 'Volatility', 'Prediction markets', 'Economic concentration', 'County executives', 'Islamic law', 'Cricket', 'Venture capital', 'Truck stops', 'Separation anxiety', 'Revolutions', 'Wolves', 'Urban planning', 'Apprenticeship', 'Hard disks', 'Local elections', 'Energy resources', 'Textiles', 'Blockchain', 'Water polo', 'Purchase orders', 'Computer industry', 'German literature', 'Digital economy', 'Gravitational waves', 'Software industry', 'Noncompliance', 'Audiology', 'Accountants', 'Memorials & monuments', 'Philosophy', 'Pole dancing', 'Polo', 'Tax planning', 'Insomnia', 'Overpopulation', 'Tolls', 'Patriarchy', 'Free will', 'Black history', 'Personality disorders', 'Policy making', 'Agricultural exports & imports', 'Inventors', 'Humanitarianism', 'Stars & galaxies', 'Performance management', 'Contempt of court', 'Dendritic cells', 'Electronic mail systems', 'Evidence', 'Surgical mesh', 'Dentistry', 'Chemical bonds', 'Magic & magicians', 'Health care', 'Sudden infant death syndrome--SIDS', 'Best management practices', 'Calligraphy', 'Postal employees', 'Profit maximization', 'Emotional disorders', 'Crystal structure', 'Aftershocks', 'User generated content', 'Construction equipment industry', 'Software services', 'Angina pectoris', 'Number systems', 'Kickboxing', 'Tracheotomy', 'Ovens & stoves', 'Colonoscopy', 'Cultural heritage', 'Invertebrates', 'Clinical outcomes', 'Insects', 'Customer relations', 'Materials management', 'Jobs credit', 'Animal cognition', 'Stock market crashes', 'Payback periods', 'Editing', 'Computer programming', 'Machine shops', 'Corrective advertising', 'Moons', 'Company automobiles', 'Honor killings', 'Circuits', 'Fetishism', 'Tongue', 'Tax revenues', 'Financial instruments', 'Boycotts', 'Class action lawsuits', 'Scholarships & fellowships', 'Guerrilla forces', 'Dogs', 'Defamation', 'Organisms', 'Employee ownership', 'Radioisotopes', 'Pasta', 'Revisionism', 'Wildcats', 'Venture capital companies', 'Beta blockers', 'Euclidean space', 'Fauvism', 'Animal assisted therapy', 'Heart attacks', 'Predatory pricing', 'Independent study', 'Genes', 'Editorials', 'Passports & visas', 'Athletic shoes', 'Competition', 'Altruism', 'Minerals', 'Wind shear', 'Malaria', 'Feeds', 'Education policy', 'Cable television industry', 'Sales territories', 'Video compression', 'Market potential', 'Sexual behavior', 'Retail sales', 'European Monetary Union', 'Police', 'Witnesses', 'Journalists', 'Fixed incomes', 'White collar crime', 'Census of Population', 'Shadow prices', 'Sex crimes', 'Bank stocks', 'Photojournalism', 'Arabs', 'Federal Reserve monetary policy', 'Iranian Americans', 'Soft drinks', 'Nitrous oxide', 'Inscriptions', 'Content management', 'Air bags', 'Bread', 'Flight simulation', 'Backpacks', 'Membranes', 'Congressional staff', 'Jet lag', 'Unmanned aerial vehicles', 'Constipation', 'Road rage', 'Sports & recreation clubs', 'Pipe organs', 'Subjectivity', 'Births', 'Nonnative species', 'Radio frequency identification', 'Corporate governance', 'Discounted cash flow', 'Kosher products', 'Sound waves', 'Financial restatements', 'Professional development', 'Circuses', 'Public housing', 'Entrepreneurial finance', 'Thrillers', 'Power', 'Petitions', 'Lupus', 'Cement plants', 'Sexual harassment', 'Layoffs', 'Budgets', 'Bank assets', 'High school baseball', 'Deutsche marks', 'Digital audio workstations', 'Safe deposit boxes', 'E coli', 'Superpowers', 'Presidents', 'Water supply', 'Land area', 'Fishing', 'Parking facilities', 'Causality', 'Chemists', 'Housework', 'Paternalism', 'Home environment', 'Brand names', 'Makerspaces', 'Universal Credit', 'Engines', 'Film studios', 'Loan originations', 'Self awareness', 'Physics', 'Rivers', 'Funeral industry', 'EC single market', 'Payroll departments', 'Contingency planning', 'Husbands', 'Environmental regulations', 'Obstruction of justice', 'Swiss banks', 'Age differences', 'Screen printing', 'Valuation', 'Conservation easements', 'Satellites', 'Foot diseases', 'Telephone banking', 'Cultural change', 'Approximation', 'Securities trading', 'French literature', 'Art markets', 'Declaratory judgments', 'Government obligations', 'Baseball', 'Trace elements', 'Sex offenders', 'Data mining', 'Heart surgery', 'Capitalism', 'Universe', 'Biometrics', 'Collateralized loan obligations', 'Mobile homes', 'Caregivers', 'Advertising revenue', 'Proxies', 'Refinancing', 'Cable cars', 'Business services', 'Ballads', 'Electric utilities', 'Public speaking', 'Bananas', 'Salvage value', 'Eye diseases', 'Election law', 'Universal life', 'Film & stage music', 'Homeowners associations', 'Photosynthesis', 'Winter', 'Poetry', 'Neolithic', 'Deoxyribonucleic acid--DNA', 'Microbiota', 'Extreme sports', 'Profit margins', 'Canning industry', 'Tax deductions', 'Wholesalers', 'Primates', 'Tempera painting', 'Frozen foods', 'Health risk assessment', 'Drama', 'Commissions', 'Evangelicalism', 'Supply chains', 'Cognition & reasoning', 'Short sales', 'Endorsements', 'Radios', 'Electronic funds transfer systems--EFTS', 'Video teleconferencing', 'Reverse engineering', 'Tax havens', 'Senescence', 'Property rights', 'Potassium', 'State parks', 'Intellectual property', 'Housewares', 'Political power', 'Divorce', 'Sugar', 'Motion sickness', 'Peanuts', 'Model airplanes', 'Diversification', 'Australian football', 'Yeast', 'Trans-Pacific Partnership Agreement', 'Talk shows', 'Actuaries', 'Skin care', 'Call centers', 'Information sharing', 'Psychological warfare', 'Deafness', 'Tear gas', 'Surgeons General', 'Observatories', 'Equality', 'Glass', 'Fuel cells', 'Legitimacy', 'Urbanization', 'Wireless communications', 'Scars', 'LPG', 'Shopping centers', 'Anti-Americanism', 'Rap music', 'Ultimate frisbee', 'Carbon dating', 'Budget deficits', 'Radio equipment', 'Fuzzy logic', 'Narratives', 'REITs', 'Historic artifacts', 'Subversive activities', 'Video recorders', 'Psychotropic drugs', 'Migraine', 'Traditions', 'Management of crises', 'Military supplies', 'Dairy farms', 'Overtime', 'General contractors', 'Military officers', 'Security systems', 'Health care networks', 'Nation building', 'Tennis', 'Enrollments', 'Virtual networks', 'Congressional elections', 'Laser printers', 'Critical path', 'User fees', 'Historical fiction', 'Estimates', 'Gauges', 'Oil fields', 'Evolution & development', 'Balance of power', 'Restricted stock', 'Historically Black Colleges & Universities', 'Gays & lesbians', 'Dyes', 'Learning', 'Volleyball', 'Pain management', 'Computer aided engineering--CAE', 'Shellfish', 'Research ships', 'Information superhighway', 'Mortgages', 'Quakers', 'Digital media', 'Work environment', 'Families & family life', 'Birthdays', 'Fiscal years', 'State visits', 'Appetite', 'Mixed media', 'Vascular endothelial growth factor', 'Academic achievement', 'Brand loyalty', 'Ivory', 'Veterans organizations', 'Staphylococcus infections', 'Planting', 'Banking', 'Herpes viruses', 'Certification', 'Toads', 'Professional sports', 'Supervisors', 'Cornea', 'Futures market', 'Book reviews', 'Military markets', 'Prosecutions', 'History', 'Ozone', 'Landscape architecture', 'Lines of credit', 'Transistors', 'Waste disposal', 'Handbags & purses', 'Astronomy', 'Independence', 'Wallcoverings', 'Harm reduction', 'Traffic congestion', 'Sales presentations', 'Military intelligence', 'Poisons', 'Beggars', 'Elk', 'Environmental impact', 'Associations', 'Podcasts', 'Printing machinery', 'Support groups', 'Nightclubs', 'Estate planning', 'Antioxidants', 'Lifeguards', 'Mitochondria', 'Anime', 'Travel & entertainment expenses', 'Federal advisory bodies', 'Military police', 'Mutual insurance companies', 'Informal economy', 'Attention deficit hyperactivity disorder', 'Fair value', 'Auditor changes', 'Plagiarism', 'Capital expenditures', 'Conjoined twins', 'Out of pocket costs', 'Shareholders equity', 'Quality of service', 'Securitization', 'Electronic filing of tax returns', 'Sexual health', 'Machinery industry', 'Competition between financial institutions', 'Frogs', 'Resins', 'Gravity', 'Cultural centers', 'Systems stability', 'Cash flow forecasting', 'Trust funds', 'Tax court decisions', 'Cannabidiol', 'Vertigo', 'Hallucinations', 'Sheriffs', 'Metabolic syndrome', 'Embryology', 'Price quotations', 'Taoism', 'Press conferences', 'Materials science', 'Ethics', 'Clearance & settlement', 'Mexican-American War', 'Charcoal', 'Amino acids', 'Tips & tipping', 'Account aggregation', 'Time use', 'Uterine cancer', 'Celiac disease', 'Meetings', 'College basketball', 'Integrative medicine', 'Diaries', 'Consumption', 'Special education', 'Cranberries', 'Enamel', 'High performance computing', 'Jazz', 'World music', 'Grills', 'Biopolymers', 'Abortion', 'Corporate treasurers', 'Fountains', 'Parliamentary elections', 'Infrared imaging systems', 'Work sheets', 'Directors', 'Gender pay gap', 'Prayer', 'Castles & palaces', 'Air pollution', 'Predictive analytics', 'Heart rate', 'Autism', 'Price wars', 'Walkways', 'Crime fiction', 'Suppliers', 'Sleds & sleighs', 'Academic grading', 'Reliquaries', 'Loyalty', 'Childrens novels', 'Air filters', 'Tuba music', 'Acne', 'Lakes', 'Colorectal cancer', 'Acrobats & acrobatics', 'Jewish life & ethics', 'Payday loans', 'Unconsciousness', 'Penguins', 'Margin requirements', 'Overdrafts', 'Assassinations & assassination attempts', 'Burglary', 'Automobile engines', 'Amicus curiae', 'Bass', 'Herbivores', 'Virtual reality', 'Testing laboratories', 'Pharmaceutical industry', 'Personnel selection', 'Chief financial officers', 'Jokes', 'Community development corporations', 'Materialism', 'Market positioning', 'Houses', 'Study abroad', 'Equity financing', 'Protestant churches', 'Passive-aggressive behavior', 'Vegetarianism', 'Beauty contests', 'Disc jockeys', 'Liposuction', 'Air conditioning', 'Overtime pay', 'Collateralized mortgage obligations', 'Emotional abuse', 'Business incubators', 'Light therapy', 'Bombings', 'Thunderstorms', 'Boer Wars', 'Motion picture festivals', 'Elasticity', 'Rentals', 'Per capita', 'Guitars', 'Flood damage', 'Enterprise search', 'Paris Agreement', 'Papal documents', 'Catholic schools', 'Fishing vessels', 'Text messaging', 'Television production', 'Dengue fever', 'Investigations', 'Territorial issues', 'Quantum dots', 'Hospitalists', 'Banking industry', 'Peers', 'Housing cooperatives', 'School dropouts', 'Rural areas', 'Gangs', 'Underemployment', 'Spasticity', 'Monetary incentives', 'Profitability', 'Sarcopenia', 'Clubs', 'Legislation', 'Figure skating', 'Mining industry', 'Credit card fees', 'Fuel economy standards', 'Generation Z', 'Futures trading', 'Recreational vehicles', 'Juvenile justice', 'Margin accounts', 'Retail stores', 'Impulsivity', 'Bisphenol A', 'Medical savings accounts', 'North American Free Trade Agreement', 'Construction costs', 'Nutrition research', 'Gated communities', 'Mars', 'Kidney stones', 'Statutes of limitations', 'Food service', 'Insurance industry', 'Floating exchange rates', 'Preemption', 'Synchronized swimming', 'Cutting tools', 'Investor behavior', 'Soft drink industry', 'Wildlife conservation', 'French language', 'Convenience foods', 'Sludge', 'Liquid crystal displays--LCDs', 'School safety', 'School lunches', 'Presbyterian churches', 'Scrap', 'Freedoms', 'Maps', 'Rodents', 'Toxicology', 'Music theory', 'Critical care', 'Nazi groups', 'Crime laboratories', 'Nervous system', 'Occupations', 'Light emitting diodes', 'Optics', 'Educational films', 'Freedom of religion', 'Gig economy', 'Customs unions', 'Helicopters', 'Osteoporosis', 'Threats', 'Compliance', 'Rheumatism', 'Emergency preparedness', 'Time capsules', 'Matter & antimatter', 'Influenza', 'Kuiper Belt', 'Mummies', 'Early modern period', 'Ice sheets', 'Price earnings ratio', 'DNA damage', 'Insect bites', 'Competitive advantage', 'Meditation', 'Pharmaceuticals', 'Abdomen', 'Picture books', 'Growth hormones', 'Personnel policies', 'Recording equipment', 'Death & dying', 'Patronage', 'Leaking of information', 'Patents', 'Weight control', 'Stock exchanges', 'Drug courts', 'Product returns', 'Conservatorship', 'Social responsibility', 'Quality standards', 'Coal industry', 'Soap', 'Criminal intent', 'Building codes', 'Native North Americans', 'Gender identity', 'Data integrity', 'Medical marijuana', 'A cappella', 'Computer engineering', 'Marching bands', 'Intoxication', 'Offshore banking', 'Deindustrialization', 'Exile', '21st century', 'Earnings per share', 'Equilibrium', 'Semantics', 'Capital distributions', 'Asset acquisitions', 'Decades', 'Connective tissue', 'Censure', 'Elbow', 'Kilns', 'Chemicals', 'Campaign expenditures', 'Respiratory therapy', 'Stress', 'Phylogenetics', 'Nuts', 'Prairie dogs', 'Desalination', 'Energy efficiency', 'Schizophrenia', 'Free blacks', 'Economic history', 'Pacifism', 'Pilgrimages', 'Payroll deductions', 'Mysteries', 'Islamic life & ethics', 'Latex allergies', 'Cigars', 'Criticism', 'Rabies', 'Tax cuts', 'Immigration policy', 'Manipulation', 'Dreams', 'Phlebotomy', 'Antennas', 'Clauses', 'Stocks', 'Walking', 'Distribution channels', 'Drunkenness', 'Consumer electronics', 'Islamization', 'Charisma', 'Murders & murder attempts', 'Tornadoes', 'Union membership', 'Salinity', 'Gestational diabetes', 'Exhibitionism', 'Fuel oil prices', 'Accident investigations', 'Censorship', 'Narcissism', 'Water quality', 'Access to information', 'Extrusion', 'Industrial safety', 'Communications systems', 'Money orders', 'Election results', 'Paleolithic', 'Modernization', 'Literary prizes', 'Famine', 'Amateurs', 'Judiciary', 'Home equity loans', 'Overqualification', 'Arabic language', 'Teaching', 'Health sciences', 'American history', 'Special effects', 'Concrete', 'Appliance industry', 'Bankruptcy laws', 'State of emergency', 'Treasury notes', 'Voting rights', 'Electronic warfare', 'Revenue management', 'Royalties', 'Muscular dystrophy', 'Melanoma', 'Endometrial cancer', 'Congressional districts', 'Chronic traumatic encephalopathy', 'Gait', 'Twins', 'Minority banks', 'Guide dogs', 'Contract proposals', 'Torture', 'School finance', 'Cost of war', 'Disabled athletes', 'Law libraries', 'Public services', 'Shelter in place', 'Industrial plant emissions', 'Blood diseases', 'Explosives', 'Spelling bees', 'Expatriate employees', 'Sport utility vehicles', 'Childlessness', 'Audited financial statements', 'Animated films', 'Logos', 'Ship accidents & safety', 'Data analysis', 'Bills', 'Furniture industry', 'Bull markets', 'Leveraged buyouts--LBO', 'Trade deficit', 'Methamphetamine', 'Potatoes', 'Ebola virus', 'Tendons', 'Missing persons', 'Metal detectors', 'Social order', 'Parents & parenting', 'Present value', 'Boating accidents & safety', 'MBA programs & graduates', 'Gun shows', 'Musical performances', 'Spacetime', 'Qualified pension plans', 'Commercial space ventures', 'Grants', 'Open data', 'Adoption', 'Labor unions', 'Bladder', 'Hormone replacement therapy', '7th century', 'Drawing', 'Pork', 'Cocoa', 'Phosphorus', 'Science teachers', 'Neoliberalism', 'Shaving & shavers', 'Sandwiches', 'Medical imaging', 'Laundries', 'Stunts & stunt performers', 'Girls clubs', 'Postpartum depression', 'Cosmetic surgery', 'Human performance', 'Crimes against humanity', 'Legal research', 'Chemical industry', 'Decision making', 'Risk assessment', 'Discounts', 'Clinical trials', 'Saxophone music', 'Uranium', 'Autobiographies', 'Housing subsidies', 'Government', 'Distracted driving', 'Foot & mouth disease', 'Hemoglobin', 'Institutionalism', 'Gross National Product--GNP', 'Mandatory retirement', 'Odors', 'Computer aided design--CAD', 'Commercial real estate loans', 'Hostage negotiations', 'Inquests', 'Government purchasing', 'Reading promotion', 'Optimization', 'Childrens art', 'Heart transplants', 'American Civil War', 'Ecosystems', 'Capital assets', 'Muslim holidays', 'Political science', 'Oil painting', 'Packaging', 'Chilled water systems', 'Income distribution', 'Green buildings', 'Dividend funds', 'Aircraft industry', 'Reading comprehension', 'Closed end funds', 'Civil engineering', 'Clay', 'Military health care', 'Filibusters', 'Credibility', 'Military vehicles', 'Conformity', 'Breeding of animals', 'Sports psychology', 'Milk production', 'Four H clubs', 'Post traumatic stress disorder', 'Gold markets', 'Liquid assets', 'Video recordings', 'High tech industries', 'National museums', 'Third party', 'Medieval period', 'Deeds', 'Microcephaly', 'Plastic surgery', 'Independence movements', 'Oil spills', 'Olive oil', 'Musical recordings', 'Body mass index', 'Climbing', 'Baby boomers', 'Cell culture', 'Adverse selection', 'Subsidiaries', 'E-books', 'Pretrial detention', 'Rehabilitation', 'Biological & chemical terrorism', 'Cave art', 'Criminology', 'Computer science', 'Industrial accidents', 'Ship hijacking', 'Crowd control', 'War', 'Probate', 'Light bulbs', 'Extraterritoriality', 'Medical diagnosis', 'Donations', 'Watercolor painting', 'Outsourcing', 'Excessive force', 'Polyphony', 'Circadian rhythm', 'Kindergarten', 'Electronics', 'Application programming interface', 'Communications satellites', 'Institutional investments', 'Television networks', 'Concentration camps', 'Forest & brush fires', 'Splitups', 'Self expression', 'Online sales', 'Policyholders', 'Mosses', 'Supernovae', 'Progress reports', 'Strategic petroleum reserve', 'Insurance coverage', 'Electricity generation', 'Eyeglasses', 'Mythology', 'Altitude', 'Locks & keys', 'Federal funding', 'Software reviews', 'Political movements', 'Gas flow', 'Billiards', 'Like kind exchange', 'Anticoagulants', 'Cultural icons', 'Natural gas industry', 'Pension costs', 'Sports agents', 'Colonies & territories', 'URLs', 'Reggae', '3-D films', 'Urban schools', 'European Currency Unit', 'Closed circuit television', 'Clearinghouses', 'Cytokines', 'Appeals', 'Mood disorders', 'River networks', 'Labor disputes', 'Cello music', 'Impoundment of appropriated funds', 'Tobacco industry', 'School violence', 'Cobalt', 'Economic rent', 'Attitude surveys', 'Carbon sequestration', 'Parapsychology', 'Sonnets', 'Human immunodeficiency virus--HIV', 'Settlements & damages', 'Used automobiles', 'Crystal oscillators', 'Consultants', 'Stranded investment', 'Strangulation', 'Amnesties', 'Private networks', 'Hemophilia', 'Pizza', 'Foreign owned US companies', 'Measles', 'Tax returns', 'Asian studies', 'Think tanks', 'Optometry', 'Income shifting', 'Throat cancer', 'Catheters', 'Civil law', 'African swine fever', 'Unemployment insurance', 'Prints and printmaking', 'Theater festivals', 'Church buildings', 'Quarantine', 'Marijuana', 'Interest rates-deposits', 'Senators', 'Dictators', 'ACT assessment', 'Latin language', 'Blackberries', 'Industrial development', 'Tragedies', 'Self help', 'Testes', 'Identity', 'Machinery', 'Bicycle racing', 'Diabetic retinopathy', 'Ranches', 'Easter', 'Fingers & toes', 'Dogsledding', 'Kidneys', 'Reading', 'Lumpectomy', 'Critics', 'Magnesium', 'Unemployment', 'Salvage', 'Exocrine glands', 'Fitness equipment', 'Consulting firms', 'Price increases', 'Campus police', 'Drug stores', 'Global health', 'Health hazards', 'Digital rights management', 'Convertibility', 'Imagination', 'International banking', 'Pageants', 'Bipartisanship', 'Epidemiology', 'Oil recovery', 'Sound design', 'Amusement parks', 'Surtax', 'Computer terminals', 'Addictive behaviors', 'Shipwrecks', 'LGBTQ community', 'Speaking', 'Clothing industry', 'Jewish art', 'Absenteeism', 'Spinal cord', 'Irrigation', 'Doppler effect', 'Social insurance numbers', 'Energy industry', 'Creative writing', 'Small business banking', 'Contractors', 'Face', 'Efficient markets', 'Vocational education', 'Peer review', 'Lutheran churches', 'Acquired immune deficiency syndrome--AIDS', 'Piracy', 'Business operations', 'Hops', 'Electronic banking', 'Aggressiveness', 'Digital maps', 'Sustainable materials', 'Tax exemptions', 'Subscription television', 'Workweeks', 'Political asylum', 'Sexual orientation discrimination', 'Green chemistry', 'Automobiles', 'Animal attacks', 'Politics', 'Auditing procedures', 'Digital music', 'Neurons', 'Pacemakers', 'Personal property', 'Disruptive innovation', 'Power supply', 'Kidney cancer', 'Racism', 'Software utilities', 'Lettuce', 'Emergency medical care', 'Shutdowns', 'Bankruptcy reorganization', 'Homebrewing', 'Tournaments & championships', 'Freedom of assembly', 'Seat belts', 'Cheating', 'Animal migration', 'Hiking', 'Audits', 'Plutonium', 'Laboratory equipment', 'Y chromosomes', 'Forward exchange', 'Volcanology', 'Ponds', 'All terrain bicycles', 'SWOT analysis', 'Bill of Rights-US', 'Wrongful discharge', 'Flowers & plants', 'Portfolio investments', 'Cognitive enhancement', 'Jewish holidays', 'Christians', 'European history', 'Religious orders', 'Drinking water', 'Adult entertainment', 'Allergens', 'Irish Americans', 'Forensic sciences', 'Auroras', 'Eyes & eyesight', 'Volume discount', 'Fathers', 'Slave trade', 'Drug formularies', 'School systems', 'Inhalers', 'Energy prices', 'Layouts', 'Xylophone music', 'Bias', 'Appointees', 'Diamonds', 'African Americans', 'Analgesics', 'Extensions', 'Student athletes', 'Unionization', 'Horse sports', 'Lead', 'Wood', 'Political transition', 'Greenhouses', 'Hispanic Americans', 'Home building', 'Geodemographics', 'Identity theft', 'Interracial relationships', 'Housing developments', 'Anxiety disorders', 'Authorship', 'STEM education', 'Slavery', 'User interface', 'Customers', 'Television production industry', 'Permeability', 'Industrial tribunals', 'Forests', 'Flag football', 'Prostheses', 'Cost of living adjustments', 'Proposals', 'Jehovahs Witnesses', 'Tax regulations', 'Dietitians', 'Bank earnings', 'Code Division Multiple Access', 'Theoretical physics', 'Human papillomavirus', 'Manures', 'Accountable care organizations', 'Patriotism', 'Passwords', 'Younger workers', 'Sport fishing', 'Nose', 'Freeware', 'Stroke', 'Language', 'Spanish language', 'Communication', 'Barbers', 'Pilot projects', 'Collectors', 'Consumer protection', 'Whooping cough', 'Ombudsman programs', 'Purchasing power parity', 'Warships', 'Land grants', 'Metals', 'Debates', 'Grammar', 'Social entrepreneurship', '5th century', 'Lifecycle funds', 'Albinism', 'Sanitation', 'Festival programming', 'Aquaculture', 'Diarrhea', 'Lobbyists', 'Research methodology', 'Extracurricular activities', 'Tools', 'Drownings', 'Deferred Action for Childhood Arrivals', 'Otherness', 'Point of sale', 'Public spaces', 'Over the counter trading', 'Loan workouts', 'Boards of directors', 'Age discrimination', 'Integrated software', 'Religious studies', 'Birth control', 'Indoor air quality', 'University students', 'Customer feedback', 'Seasons', 'Divestments', 'Process controls', 'Globalization', 'Overhead costs', 'Preschool education', 'Trustees', 'Romantic music (Classical)', 'Summer', 'Asteroids', 'Taxonomy', 'PBX', 'Heavy metals', 'Funeral homes', 'Convention centers', 'September 11 terrorist attacks-2001', 'Nausea', 'Goal setting', 'Gambling industry', 'Ghosts', 'Household utilities', 'Towns', 'Metamorphic rocks', 'Criminal sentences', 'Heating', 'Paralysis', 'Shipping industry', 'Market penetration', 'Financial Services Modernization Act 1999-US', 'Oil and gas leases', 'Mass transit', 'Tolerance', 'Moral injury', 'Consumer behavior', 'Ultraviolet radiation', 'Cats', 'Police training', 'Marketing', 'Hematology', 'Vans', 'Chat rooms', 'Export controls', 'Historic buildings & sites', 'Shareholder voting', 'Pawn shops', 'Self control', 'Revenue sharing', 'Expungement', 'Citizens', 'Salads', 'Neutron stars', 'Ransomware', 'Educational vouchers', 'Faith', 'Nuclear engineering', 'Railroad accidents & safety', 'Human remains', 'Equity capital', 'Access control', 'Child development', 'Flight attendants', 'Advertising campaigns', 'Cookies', 'Food supply', 'Industrial wastes', 'Tax allocation', 'Integrated circuits', 'Epigenetics', 'Poker', 'Synthetic biology', 'Charged particles', 'Dadaism', 'Embolization', 'Hyperglycemia', 'Book value', 'Regulation of financial institutions', 'State laws', 'Abandoned mines', 'Pastels', 'Privacy', 'Veterinary medicine', 'User groups', 'Chain stores', 'Solar cycle', 'Blood tests', 'Picketing', 'Municipal finance', 'Aerospace materials', 'Chinese exclusion laws', 'Angel investors', 'Fair Housing Act 1968-US', 'Yodeling', 'Automobile sales', 'Tax reform', 'Handedness', 'Presidential vetoes', 'Cameras', 'Service centers', 'Preferences', 'Journals', 'Play', 'Natural gas utilities', 'State government', 'Project management', 'Alopecia', 'Executive orders', 'Hindus', 'Mesothelioma', 'Armor', 'Shakespeare plays', 'Tapestry', 'Economic crisis', 'Opening hours', 'Kayaking', 'School districts', 'Pilgrims', 'Nongovernmental organizations--NGOs', 'Space exploration', 'FDA approval', 'Nanoparticles', 'Blueberries', 'Record labels', 'Airline scheduling', 'Bending stresses', 'Cyberbullying', 'Book clubs', 'Codes', 'Securities analysis', 'Tax audits', 'Religious broadcasting', 'Equity', 'Stamps', 'Religious persecution', 'Working parents', 'Mercury', 'Soul music', 'Bahaism', 'Bathhouses', 'Art therapy', 'Mid cap investments', 'Insider trading', 'Community Reinvestment Act 1977-US', 'Malware', 'Ice shelves', 'Physician assistants', 'Windmills', 'Handball', 'Education reform', 'Comparable worth', 'States rights', 'Drug testing', 'Role models', 'Heat exchangers', 'Art galleries & museums', 'Product safety', 'Compound interest', 'Perfumes', 'Transportation services', 'Shift work', 'Property values', 'Accounts payable', 'Trade secrets', 'Acids', 'Fashion models', 'Oceans', 'Hostile takeovers', 'Agricultural production', 'Knowledge', 'Commercial real estate', 'Imprisonment', 'Electronic waste', 'Common Agricultural Policy', 'Audiobooks', 'Platinum', 'Manual workers', 'Sex education', 'Family law', 'Coroners', 'Cafeterias', 'Artists studios', 'Travel', 'Oratorios', 'Sepsis', 'Economic impact', 'Lotteries', 'Labor shortages', 'Energy management', 'Waiting period', 'Company publications', 'Charities', '13th century', 'Securities markets', 'Folk music', 'Single family', 'Due diligence', 'Preventive maintenance', 'Polar bears', 'Hepatitis B', 'Big business', 'Air transportation industry', 'Empowerment', 'Student retention', 'Segregated funds', 'Clitoris', 'Ballet companies', 'Export import banks', 'Office space', 'Ramadan', 'Rotavirus', 'Employers', 'Forgery', 'Church & state', 'Shipyards', 'Mentors', 'Professional relationships', 'Bird migration', 'Concrete floors', 'Debt exchanges', "Trompe l'oeil", 'Mosquitoes', 'Plantations', 'Clean technology', 'Carbon dioxide', 'Refineries', 'Thin films', 'Appliances', 'Stock brokers', 'Provincial elections', 'Modern art', 'Franchising', 'Consumer spending', 'Essays', 'Independent counsels', 'Photonics', 'Patient satisfaction', 'Natural language', 'Manga', 'Day spas', 'Osmosis', 'Flour', 'Transition economies', 'Scanners', 'Phonics', 'Tax services', 'French Canadians', 'Deficit financing', 'Gastroenterology', 'Racial differences', 'Combatants', 'Oppression', 'Lithium', 'Leasing', 'Additive manufacturing', 'Active listening', 'Foreign exchange markets', 'Nonviolence', 'Stock rights', 'Barley', 'Gas industry', 'Natural gas', 'Clinics', 'Instant messaging', 'Riders', 'Food processing industry', 'Acquittals & mistrials', 'Social exclusion', 'Participation', 'Habitual offenders', 'Hunger', 'Maritime policy', 'Christmas', 'Financial leverage', 'Crohns disease', 'Bank portfolios', 'Courthouses', 'File sharing', 'After school programs', 'Levees & battures', 'Spreadsheets', 'Islamic countries', 'Photography', 'Human geography', 'Illiteracy', 'Mirrors', 'Science history', 'Electrodes', 'Respiratory syncytial virus', 'Bolshevism', 'Cocoa industry', 'Bank service charges', 'Peritoneal dialysis', 'Neighborhood blight', 'Skills', 'Quarks', 'Affirmative action', 'Electronic music', 'Mobile content', 'Pollutants', 'Singing', 'Birthing centers', 'Regulation', 'Fish production', 'Condiments', 'Psychoanalysis', 'Poaching', 'Parole boards', 'Escheat laws', 'Piers', 'Canadian dollar', 'Statins', 'Soybeans', 'Bank statements', 'Outbuildings', 'Constitutional conventions', 'Magazines', 'Grocery industry', 'Health facilities', 'Arachnids', 'Native rights', 'Losses', 'Asian Americans', 'Stamp duties', 'Editors', 'Securities lending', 'School environment', 'Stents', 'Withholding taxes', 'Health risks', 'Byzantine civilization', 'Efficiency', 'Computer crime', 'Good & evil', 'National identity', 'Conscientious objectors', 'Solar physics', 'Government investigations', 'Software engineering', 'Bars', 'Estimated taxes', 'Government securities', 'Dopamine', 'Womens suffrage', 'Threat assessment', 'Wrist', 'Manslaughter', 'Insurance applications', 'Legal ethics', 'Chief operating officers', 'French as a second language', 'Engineering', 'Timesharing', 'Packaged goods', 'Encephalitis', 'Cattle', 'Contactless payment', 'Mission statements', 'Tax treaties', 'Satellite television', 'Debt cancellation', 'Economic life', 'Plasma', 'Badminton', 'Clothing', 'Aerospace engineering', 'Corporate culture', 'Car pools', 'Lighting', 'American Depositary Receipts', 'Military aid', 'Mark to market accounting', 'Closely held corporations', 'Spanish culture', 'Psoriasis', 'Epilepsy', 'Validity', 'God', 'Lyricists', 'Music industry', 'Mechanics', 'Performance evaluation', 'International markets', 'Propensity to save', 'Automation', 'Impeachment', 'Generic drugs', 'Multinational corporations', 'Capital structure', 'Correspondent banks', 'Mortgage backed securities', 'By products', 'Urticaria', 'Fatty acids', 'Product life cycle', 'Womens literature', 'Bollywood films', 'Polyethylene', 'Provinces', 'Health maintenance organizations--HMOs', 'Underground construction', 'Visualization', 'Skateboarding', 'False advertising', 'Theater companies', 'Child mortality', 'Maintenance costs', 'Civil rights movements', 'Privacy Act 1974-US', 'Biochemistry', 'Pandas', 'Convenience', 'Office furniture', 'Light rail transportation', 'Massage parlors', 'Gender equity', 'Protestant Reformation', 'Structured notes', 'Steam power', 'Deadly force', 'Reputations', 'Testicular cancer', 'Deposition', 'Methadone', 'Misappropriation of funds', 'Documentation', 'Sleep apnea', 'Entrepreneurship', 'Clinical psychology', 'Queries', 'Stalinism', 'Catastrophes', 'Molybdenum', 'Airline industry', 'Cash flow', 'Seasonal affective disorder', 'Blue collar workers', 'Business income', 'Social activism', 'Charitable trusts', 'Childrens rights', 'Military withdrawals', 'Retailing industry', 'Cloud computing', 'Canola', 'Isolationism', 'Declaration of Independence-US', 'Mining', 'Thyroid gland', 'Foundries', 'Environmental policy', 'User needs', 'All terrain vehicles', 'Rites of passage', 'Raids', 'Rents', 'Pricing policies', 'Trade promotion', 'Child & adolescent psychiatry', 'Economic opportunities', 'Personal identification numbers', 'Found objects', 'Enzymes', 'Medical treatment', 'Chlorofluorocarbons', 'Bear markets', 'Public schools', 'Cellulose', 'British culture', 'Academic achievement gaps', 'Work life balance', 'Renewable resources', 'Hispanics', 'Mortgage companies', 'Bioinformatics', 'Opera', 'Writedowns', 'Strategic management', 'Sports injuries', 'Gambling', 'Problems', 'Flexible hours', 'Game theory', 'Majority stockholders', 'Help wanted ads', 'Nonaudit services', 'Winter sports', 'Expert witness testimony', 'Quality of education', 'Bicycling', 'Proteins', 'Energy', 'Diagnostic tests', 'Breakfast foods', 'Military training', 'Employee benefits', 'Conceptual art', 'Groundwater', 'Aromatherapy', 'Corporate responsibility', 'Chinese history', 'Flexible packaging', 'Records & achievements', 'Fences', 'Royalty', 'Absentee voting', 'Metastasis', 'Hemodialysis', 'National monuments', 'Mysticism', 'Watersheds', 'Teenage pregnancy', 'Social capital', 'Passengers', 'Animal populations', 'Seismology', 'Counterfeiting', 'Bassists', 'Asbestos', 'Child sexual abuse', 'Software', 'Invasions', 'Fuel taxes', 'Automobile theft', 'Superheroes', 'Amnesia', 'Record producers', 'Defibrillators', 'Electronic monitoring', 'Corrosion', 'Accounting changes', 'Zero tolerance', 'Steel production', 'Buddhists', 'Atoms & subatomic particles', 'Graduation rate', 'National parks', 'Private members bills', 'Victimization', 'Joint stock companies', 'Liquor laws & regulations', 'Black swan event', 'Dehydrogenases', 'Local economy', 'Environmental studies', 'Chromatography', 'Auditing', 'Book challenges', 'Mitochondrial DNA', 'Textile design', 'Heart', 'Trust companies', 'Therapists', 'Student organizations', 'Liver', 'Economic theory', 'Pop-up retail', 'Regional stocks', 'Boat racing', 'Geothermal power', 'Military service', 'Law', 'Bureaucracy', 'Torah', 'American Revolution', 'Professional recruitment', 'Newsletters', 'Electric power', 'Meat processing', 'Patrol cars', 'Estates', 'Speculative fiction', 'City ordinances', 'Market research', 'Freedom of movement', 'Line managers', 'Broadway theater', 'Collaboration', 'Slang', 'Surface water', 'Economic value added', 'Chemical filters', 'Head & neck cancer', 'Job fairs', 'Short stories', 'Holidays & special occasions', 'Annexation', 'Point of care testing', 'Brain damage', 'Security personnel', 'Kinesiology', 'Marxism', 'Farm machinery', 'Great Depression', 'Navigation systems', 'Natural gas prices', 'Drug overdose', 'Repurchase agreements', 'Commercial space', 'Legal advocacy', 'Evidence-based medicine', 'Antiretroviral drugs', 'Personal trainers', 'Sentiment analysis', 'Transitional justice', 'Redistricting', 'Bans', 'Electrolytes', 'Bioethics', 'Land-grant colleges', 'Sustainability', 'Zinc', 'Antitrust', 'Inheritances', 'Chinese medicine', 'Hydration', 'Hardship', 'Overproduction', 'Foreign aid', 'Shareholder approval', 'Surcharges', 'Jet fuel prices', 'Manufacturers', 'Vandalism', 'Thanksgiving', 'Cognitive psychology', 'Transfer payments', 'Hypertension', 'Data processing', 'Radar systems', 'Youth organizations', 'Grinding tools', 'Friendly fire', 'Reunions', 'Credit bureaus', 'Counseling', 'Storms', 'Personal relationships', 'Nuclear fission', 'Linens', 'Prohibition era', 'Cardiopulmonary resuscitation--CPR', 'Rate setting', 'Learning disabilities', 'Trade negotiation', 'Hierarchies', 'Paramilitary groups', 'International trade', 'Drugs & sports', 'Identification documents', 'Progressive taxes', 'Scale and proportion', 'Cardiology', 'Deforestation', 'Rubber', 'Partnership agreements', 'Oil shale', 'Fairs & exhibitions', 'Market research firms', 'Males', 'Social services', 'Telecommunications towers', 'Vomiting', 'FISC', 'Firings', 'Prison overcrowding', 'Coaching', 'Police reports', 'Military technology', 'Social justice', 'Pet supplies', 'Wireless access points', 'Crowdfunding', 'Hypothalamus', 'Time management', 'Animals', 'Multilateralism', 'Military weapons', 'Digital audio players', 'Ratings & rankings', 'Stop & frisk', 'Mining engineering', 'Mobile communications networks', 'Tuna', 'Education', 'High speed', 'Governmental reform', 'Commercial banks', 'Personal archives', 'Internal controls', 'Armed forces', 'Impressionism', 'Fetuses', 'Capital costs', 'IPTV', 'Beer', 'Public opinion', 'Surfing', 'Curfews', 'Washers & dryers', 'Constitutional review', 'Safes', 'Artistic directors', 'Writers', 'Snow', 'Set top boxes', 'Gene expression', 'Grocery stores', 'Ships', 'Immunity from prosecution', 'Animal euthanasia', 'Collections', 'Medical supplies', 'Automotive engineering', 'Federal regulation', 'Warehouses', 'Tenure', 'Mechanical engineering', 'Furniture stores', 'Stabbings', 'Exoticism', 'Digital divide', 'Historians', 'International', 'Endocrine system', 'Self interest', 'Nonprofit organizations', 'Body composition', 'Political campaigns', 'Scandals', 'Computer peripherals', 'Selfishness', 'Employment discrimination', 'Nobility', 'Poultry', 'Asbestos industry', 'Flat rates', 'Corn', 'Olives', 'Hormones', 'Portable computers', 'Vetoes', 'Popular culture', 'Religious figures', 'Polls & surveys', 'Otolaryngology', 'Onsite', 'Hypnotherapy', 'Swine flu', 'Egalitarianism', 'Costume design', 'SAT assessment', 'Photographic equipment & supplies', 'Volunteers', 'Health insurance', 'Consumer Confidence Index', 'Cross-dressers', 'Bats', 'Supermarkets', 'Blues music', 'Horror films', 'Profanity', 'Peer tutoring', 'Artists', 'Global positioning systems--GPS', 'Happiness', 'Environmentalism', 'Revenue recognition', 'Agricultural lending', 'Career advancement', 'Employees', 'Constituents', 'Safety & soundness', 'Easements', 'Nuclear security', 'Anthologies', 'Heresy', 'Populism', 'Lifestyles', 'Cerebral palsy', 'Fishing zones', 'Phillips curve', 'Natural products', 'Tax avoidance', 'Magnetic fields', 'Punitive damages', 'aphair%2Btest%2Baphair', 'Voice messaging systems', 'Relational data bases', 'Orthopedics', 'Compact disc players', 'Regional government', 'Researchers', 'Industrial markets', 'Methodist churches', 'Trust', '15th century', 'Abandonments', 'Tax increases', 'Inhibitor drugs', 'Soil sciences', 'Religious education', 'Raises', 'Military cemeteries and funerals', 'Business closings', 'Bond portfolios', 'Project finance', 'Empathy', 'Roman civilization', 'Film musicals', 'Cadmium', 'Reiki', 'Air rights', 'Civilians', 'Digital preservation', 'Criminalization', 'Orchestras', 'Yom Kippur', 'Nuclear tests', 'Nihilism', 'Return on assets', 'Television programming', 'Shareholder activism', 'Confidence', 'Power marketers', 'Lifesaving', 'Retreats', 'Daylight saving time', 'Energy infrastructure security', 'Jaw', 'Market equilibrium', 'Mobile advertising', 'Baby strollers', 'Toilet training', 'Automobile dealers', 'Christianity', 'Subtitles & subtitling', 'Engineers', 'Educational attainment', 'Online tutorials', 'Coaches & managers', 'Toys', 'Christmas music', 'Processing speed', 'Externality', 'Civil defense', 'Torts', 'Management buyouts', 'Fatigue', 'World War I', 'Federal government', 'Narcotics', 'Reimbursement', 'Deposit insurance', 'Handicapped accessibility', 'Vaping', 'Radioactive wastes', 'Sales quotas', 'Income taxes', 'Copper', 'Bribery', 'Tax courts', 'Foreign exchange rates', 'Liver transplants', 'Lasers', 'Tire industry', 'Heaters', 'Comebacks', 'Oncology', 'Catfish', 'Luxury automobiles', 'Ammunition', 'Buddhism', 'Online securities trading', 'Nonlethal weapons', 'Reagents', 'Genetically modified organisms', 'Hostels', 'Military medicine', 'Trials', 'Lung cancer', 'Fruit juices', 'Talmud', 'Anorexia', 'Surgical outcomes', 'European integration', 'Proxy statements', 'College baseball', 'Human capital', 'Lipstick', 'Severance pay', 'Health savings accounts', 'Price fixing', 'Symmetry', 'Shareholder derivative suits', 'Hockey', 'Japanese history', 'Capital requirements', 'Galvanized steel', 'Turner Prize', 'Dioramas', 'Professional soccer', 'Crowdsourcing', 'Catering', 'Travel literature', 'Supranationalism', 'Electron tubes', 'Whistleblowing', 'Clergy', 'Web browsers', 'Bail', 'Social policy', 'Curators', 'Tobacco', 'Catatonia', 'Modern pentathlon', 'Psychic phenomena', 'Chamber music', 'Corporate debt', 'Realism', 'Zoology', 'Dividend policy', 'Russian history', 'Unfair labor practices', 'Spinoffs', 'Scope of practice', 'Sailing & sailboats', 'Fireplaces', 'Speedy trial', 'Financial performance', 'Fans (Aficionados)', 'Goodwill', 'Probability', 'Weightlifting', 'Social unrest', 'Knives', 'Vietnam War', 'Information professionals', 'Exhibits', 'Peer relationships', 'Pathology', 'Multitasking', 'Bond funds', 'Small business loans', 'Vagina', 'Thrombosis', 'Endodontics', 'Propaganda', 'Lying', 'Weapons testing', 'Music festivals', 'Hot flash', 'Premature babies', 'Cashmere', 'Exigent circumstances', 'Publishing', 'Friction', 'Industrial space', 'Foreign labor', 'Insurance adjusters', 'Industrial equipment', 'Purchasing', 'Birth rate', 'Infomercials', 'Online gambling', 'Facial recognition technology', 'Student loans', 'Waste materials', 'Territorial transfers', 'Single persons', 'Speed limits', 'Organic farming', 'Automobile insurance', 'Yachts', 'Open meeting laws', 'African art', 'Oral administration', 'Crime', 'Special purpose acquisition companies', 'Beautification projects', 'Cyberterrorism', 'Metal workers', 'Bank reserves', 'Fishing industry', 'Bar codes', 'Reverse discrimination', 'Word processing', 'Islamophobia', 'Submarines', 'Nitrogen dioxide', 'Bipartisan Campaign Reform Act 2002-US', 'Designers', 'Graffiti', 'Nursing homes', 'Paleoclimate science', 'Automobile parking', 'Awards & honors', 'Water pollution', 'Accelerated death benefits', 'Marinas', 'Car washes', 'Data transmission', 'Bank examinations', 'Child abuse & neglect', 'Chloride', 'Regulatory agencies', 'Bilingual education', 'Research', 'Foreign banks in US', 'Entrepreneurs', 'Advisors', 'Pardons', 'Medical personnel', 'Sand & gravel', 'Automobile registration', 'Stains & staining', 'High school football', 'Refugee camps', 'Citrus fruits', 'Surveillance of citizens', 'Crocodiles', 'Gasoline prices', 'Sea transport', 'Leadership training', 'Privatization', 'Forfeitures', 'Monsters', 'Comedians', 'Folklore', 'Social skills', 'Meals', 'Team handball', 'Jewelry', 'Memberships', 'Quantum field theory', 'Constitutional amendments', 'Tobacco smoke', 'Biotechnology industry', 'Catholic churches', 'Severe acute respiratory syndrome', 'Fear & phobias', 'Airplane engines', 'Single sex education', 'Coffee industry', 'Illustrators', 'Sports executives', 'Organic chemicals', 'Radioactivity', 'Financial executives', 'Visual impairment', 'Prosperity', 'Safety devices', 'Book fairs', 'Iodine', 'Corporate profiles', 'Interest groups', 'Treasuries', 'Conservatism', 'Seminars', 'Fistula', 'Miranda rights', 'Testosterone', 'Alumni associations', 'Hippies', 'Drive by shootings', 'Criminal law', 'Telecommunications systems', 'Luggage', 'Endocrinology', 'Inheritance taxes', 'Bingo', 'Sarcophagi', 'Government bonds', 'Insurance policies', 'Bedding', 'Textile industry', 'Artificial reefs', 'Economic justification', 'Ingredients', 'Light pollution', 'Basements & cellars', 'Wire fraud', 'Contract negotiations', 'Counterrevolution', 'Currencies', 'Political ethics', 'Students', 'Corporate liability', 'Teamwork', 'Aspirin', 'Statistics', 'Silver', 'Intersectionality', 'Palliative care', 'Billboards', 'Robots', 'Asian literature', 'Embargoes & blockades', 'Substance abuse treatment', 'Classical literature', 'Wagner Act 1935-US', 'Ultrasonic imaging', 'Kitchens', 'Palladium', 'Digital libraries', 'Religious identity', 'Heavy construction', 'Business cycles', 'Verbal communication', 'Public policy', 'Earth Day', 'Antibodies', 'Advertising restrictions', 'Civilization', 'Health Insurance Portability & Accountability Act 1996-US', 'Peptides', 'Wigs', 'Furniture', 'Euro', 'Nuclear weapons', 'Prices', 'Information sources', 'Developing countries--LDCs', 'Swimming pools', 'Sodium', 'Seasonal markets', 'Liquor', 'Domestic terrorism', 'Theft', 'Emissions trading', 'Gloves', 'Reverse mortgages', 'Housing policy', 'Velocity', 'Foreign investment', 'Scale models', 'Executives', 'Wireless carriers', 'Nuclear physics', 'Psychologists', 'Home invasion', 'Savings accounts', 'Interventionism', 'Charter schools', 'Uninsured people', 'Vinegar', 'Kidney diseases', 'Virology', 'Microorganisms', 'Bronchitis', 'Asian students', 'Independent bookstores', 'Chlamydia', 'Paternity', 'Veterans', 'Physical therapists', 'Steel industry', 'Product management', 'Job creation', 'Partnering', 'Books', 'Wheels', 'High definition television--HDTV', 'Mental health', 'Classification', 'Psychiatrists', 'Clawback', 'Collective bargaining', 'Liquor stores', 'Oldest old people', 'Handicrafts', 'Sexism', 'Aerial surveying', 'Media coverage', 'Clearing banks', 'Noncitizens', 'Enterprise risk management', 'Academic probation', 'Industrial engineering', 'Field trips', 'Beef', 'Morphology', 'Head injuries', 'COVID-19', 'Railroads', 'Operating companies', 'Going public', 'Flags', 'Remarriage', 'Cultural anthropology', 'Advantages', 'Middle school students', 'Hebrew language', 'Gluten', 'Urban farming', 'Seniority', 'Sturgeon', 'Delirium', 'Digital marketing', 'Salmon', 'College students', 'Taxable income', 'Back up systems', 'Religious organizations', 'Self service', 'Diesel engines', 'Tombs', 'Genocide', 'Interfaces', 'Orbits', 'Retirees', 'Dating services', 'Television sets', 'Energy storage', 'Grandparents', 'Nuclear power plants', 'Technology transfer', 'Energy modeling', 'Hemp', 'Sons', 'Festivals', 'Sheet music', 'Gross Domestic Product--GDP', 'Conglomerates', 'Medical referrals', 'Warm up (exercise)', 'Occupational stress', 'Pension plans', 'Chief information officers', 'Life insurance companies', 'Partial Birth Abortion Ban Act 2003-US', 'Uniforms', 'Collisions', 'Moon', 'Indictments', '18th century', 'Peronism', 'Mortgage banks', 'Elections', 'Mouth', 'Italian Americans', 'Noise', 'Numbers', 'Water sports', 'Athletic drafts & trades', 'Leptospirosis', 'Campaign contributions', 'Holocaust', 'Investments', 'Magma', 'Employment', 'Cesarean section', 'Shopping carts', 'Biodiesel fuels', 'Feces', 'Baldness', 'Miracles', 'Information science', 'Aircraft hijacking', 'Cremation', 'Capital leases', 'Home meal replacement', 'Scholars', 'Macroeconomics', 'Childbirth & labor', 'Beverages', 'Retail banking', 'Group health insurance', 'Data compression', 'Security portfolios', 'Disability', 'African American studies', 'Personal information', 'Otology', 'Suburban areas', 'Professional hockey', 'Chronic obstructive pulmonary disease', 'Steel pipes', 'Operations management', 'Night vision', 'Contemporary art', 'Certificates of deposit', 'Height', 'Knee', 'Nuns', 'Nitrogen', 'Warrants', 'Apathy', 'Appraisals', 'Vortices', 'Common Security and Defence Policy', 'Foreign Intelligence Surveillance Act 1978-US', 'Ephemera', 'Mathematics education', 'Environmental quality', 'Truth in advertising', 'Bone density', 'National security', 'Myopia', 'Dispersal', 'Japanese language', 'Art exhibits', 'Candidates', 'Prime time', 'Economics education', 'Electronic government', 'Gifts', 'Gender differences', 'Investment banking', 'Investment Canada Act', 'Soil erosion', 'Child care', 'Islamism', 'Masculinity', 'Business forecasts', 'Cruises', 'Spectrum allocation', 'Mind body relationship', 'Industrial plants', 'Mentoring programs', 'Native languages', 'Nutrients', 'Air cleanliness', 'Political correctness', 'Drones', 'Consignment buying', 'Vacancies', 'House arrest', 'Interferon', 'Tender offers', 'Sharing economy', 'Migration', 'Inventions', 'Official misconduct', 'Noise pollution', 'Intelligence tests', 'Carcinogens', 'Forensic pathology', 'Wildlife trade', 'Protocol', 'Orthopedic apparatus', 'Security clearances', 'Transcranial magnetic stimulation', 'Voodoo', 'Traffic accidents & safety', 'Flooring', 'Hepatitis C', 'Subsidies', 'Chronic fatigue syndrome', 'Theory of relativity', 'Culinary schools', 'Humane societies', 'Jet engines', 'Animal bites', 'Bank acquisitions & mergers', 'Laboratory animals', 'Chocolate', 'Money laundering', 'Prisons', 'Ocean currents', 'Television miniseries', 'Coal-fired power plants', 'Teams', 'Core curriculum', 'Foreign investments in the US', 'Westerns', 'Television stations', 'Beaches', 'Petroleum', 'Thyroid cancer', 'Land mines', 'Gastrointestinal surgery', 'Hegemony', 'User experience', 'Internal customers', 'National service', 'Law schools', 'Nuclear winter', 'Pulitzer prizes', 'Electronic newspapers', 'Collection services', 'Oil wells', 'Balance sheets', 'Fairy tales', 'Letters', 'Management training', 'Conflicts of interest', 'Girls', 'Conspiracy', 'Product differentiation', 'Small intestine', 'Weaving', 'Recidivism', 'Ovarian cancer', 'Dry cleaners', 'Social support', 'Televised trials', 'Space stations', 'Urban history', 'Golf', 'Knighthood', 'Organized crime', 'Underwriting', 'Rock music', 'Fixed rates', 'Sanitizers', 'Anger management', 'Supreme Court decisions', 'Genomics', 'Credit Card Accountability Responsibility & Disclosure Act 2009-US', 'Peace', 'Cabaret', 'Shipbuilding', 'Quality', 'Sediments', 'Russian language', 'Business insurance', 'Regional banks', 'Investment advisors', 'School employees', 'Metal fatigue', 'Laity', 'Financial disclosure', 'Joint operating agreements--JOA', 'Online voting', 'Flood control', 'Corporate credit unions', 'Self employment', 'Explosions', 'Receivables', 'Spelling', 'New Left', 'Executive agreements', 'Appraisers', 'Vouchers', 'Bottled water', 'Alternative minimum tax', 'Cultural Revolution', 'Right of privacy', 'Community service', 'Financial counseling', 'Stress response', 'Penicillin', 'Income inequality', 'Urinary incontinence', 'Digital video', 'School superintendents', 'Pewter', 'Electric instruments', 'Choreography', 'Market timing', 'Archives & records', 'Hydrogen peroxide', 'Cost estimates', 'Shelters', 'Sales culture', 'Taxpayers', 'Celibacy', 'Brownfields', 'DVD', 'Forest management', 'Transportation industry', 'Asphalt pavements', 'Monopolies', 'Nosocomial infections', 'Credit unions', 'World War II', 'Harmonica music', 'Soccer', 'Meteorological satellites', 'Smart houses', 'Athletic recruitment', 'Soils', 'Military draft', 'Securities regulations', 'Anthropology', 'Monasteries', 'Military bases', 'Parliaments', 'Radical groups', 'Shareholder meetings', 'Jupiter', 'Teacher education', 'Bank technology', 'Water markets', 'Identity politics', 'Covert operations', 'Radio stations', 'Hip hop music', 'Estate administration', 'Aperture', 'Garnishment', 'Precision medicine', 'Short term', 'Food safety', 'Graphic novels', 'Russian culture', 'Biographies', 'Product recalls', 'Canadians', 'Oceanography', 'Philanthropy', 'Demography', 'Production costs', 'Rafting', 'Punk rock music', 'US exports', 'Generation X', 'Politicians', 'Vectors (Biology)', 'Beards & mustaches', 'Greenhouse effect', 'Dark energy', 'Dietary supplements', 'Cartoonists', 'Airline code sharing', 'Race relations', 'Iraq War-2003', 'Psychiatry', 'Postwar reconstruction', 'Black students', 'Restraint of trade', 'Crustaceans', 'Digital currencies', 'Balanced Budget & Emergency Deficit Control Act 1985-US', 'Building construction', 'Options markets', 'Economic policy', 'Surgical apparatus & instruments', 'Dairy industry', 'Leasebacks', 'Astrophysics', 'Calendars', 'Temperature', 'User services', 'Industrial gases', 'Government executives', 'Teachers', 'Trauma care', 'Abstract expressionism', 'Commodity futures', 'Cervical cancer', 'Claims processing', 'Projected images', 'Hyperventilation', 'Leprosy', 'Secularism', 'Cooperation', 'Superconducting supercolliders', 'Memorial Day', 'Endowment', 'Country music', 'Deadlines', 'Temporary housing', 'Satellite communications', 'Brochures', 'Voter registration', 'Veterinarians', 'Outdoor advertising', 'Shrines', 'Protective coatings', 'Political parties', 'Solar system', 'Bicycles', 'Analytical chemistry', 'Congressional investigations', 'Concertos', 'Science fiction & fantasy', 'Growth models', 'Beneficiaries', 'Disability insurance', 'Herbs', 'Charters', 'Farmworkers', 'Questionnaires', 'Risk retention', 'Confiscation', 'Bargaining', 'Pathogens', 'Journalism schools', 'Accounting procedures', 'Earth science', 'Mortality', 'Competency tests', 'Fuel cell vehicles', 'Dictionaries', 'Labor economics', 'Hispanic students', 'Energy research', 'Simplification', 'Market prices', 'Parasitic diseases', 'Nationalization', 'Pancreatic cancer', 'Business valuation', 'Sales promotions', 'Garlic', 'Farmers', 'Retirement', 'Cross country running', 'Heimlich maneuver', 'Bank accounts', 'Brain', 'International accounting standards', 'Modern history', 'Rain', 'Wetlands', 'Chinese Americans', 'Medical students', 'Pneumatics', 'Sinuses', 'Career development planning', 'Foreign language learning', 'Newborn babies', 'Carbon footprint', 'Connectors', 'Lungs', 'Vegetable oils', 'Nonconforming loans', 'Offshore oil exploration & development', 'Energy drinks', 'Screenplays', 'Fraternal organizations', 'Avian flu', 'Yen', 'Traffic police', 'Transportation terminals', 'Friendship', 'Community support', 'International courts', 'Maritime security', 'Design thinking', 'Education management organizations', 'Olympic games', 'Business failures', 'Coasts', 'Gasoline', 'Hydrogenation', 'Office parks', 'Regulatory approval', 'Inuktitut language', 'Capital gains', 'Fashion shows', 'School buses', 'Endocrine therapy', 'Religious conversion', 'Product lines', 'Polish Americans', 'Sensors', 'Library associations', 'Electronic trading systems', 'Intelligence services', 'Political activism', 'Shootings', 'Deepwater drilling', 'Glasnost', 'Contact tracing', 'Religion & politics', 'Market segments', 'Apples', 'Death threats', 'Parasites', 'High yield investments', 'Affinity groups', 'Depreciation', 'Journalism', 'Mathematical problems', 'Real estate sales', 'Portfolio performance', 'Orphanages', 'Hepatitis', 'State employees', 'Autographs', 'Corporate income tax', 'Virtual private networks', 'Accounting policies', 'Independent films', 'Political prisoners', 'Novels', 'Exclusion', 'Telephone answering services', 'Confidential relationships', 'Pediatrics', 'Aggregates', 'Rubles', 'Milk', 'Discount coupons', 'Siblings', 'Tourette syndrome', 'Franchises', 'Yard waste', 'Financial reporting', 'Microprocessors', 'Socialization', 'Forest products', 'Magnetic tape', 'Northern communities', 'Attitudes', 'Range of motion', 'Sewer systems', 'Camps', 'Lightning', 'Vice Presidents', 'Insolvency', 'Voting machines', 'Leases', 'Latin music', 'Religious discrimination', 'Aluminum alloys', 'Attorneys general', 'Automobile fleets', 'Self regulation', 'Sulfur', 'Foreign trade zones', 'Vendors', 'Trade finance', 'Wireless roaming', 'Elder care', 'Enterprise zones', 'Bioengineering', 'Currency revaluation', 'Land use planning', 'Egyptian civilization', 'Flame retardants', 'Argentine culture', 'Remote computing', 'Standard deviation', 'Credit risk', 'Business writing', 'Art', 'Islands', 'Insulin', 'Petrochemicals industry', 'Energy shortages', 'Adventure', 'Ovaries', 'Real time', 'Permits', 'Golf shoes', 'Allergies', 'Graduations & commencements', 'Convenience stores', 'Conveyor lines', 'Animal vaccines', 'Smell', 'Microelectromechanical systems', 'Alcoholic beverages', 'Champagne', 'Savings banks', 'Gas turbines', 'Generators', 'Cryogenic engineering', 'Sabotage', 'Obsessive compulsive disorder', 'Popular music', 'Random access memory', 'Municipal government', 'Discretionary income', 'Ice', 'Textile research', 'Contempt of Congress', 'Metaphor', 'Cohabitation', 'Hymns', 'Nudity', 'Asset management', 'Surfactants', 'Jazz ensembles', 'Beverage industry', 'Consent', 'Religious schools', 'Honey', 'Truancy', 'Babies', 'Delinquency', 'Fire extinguishers', 'Capital stock', 'Referendums', 'German language', 'High speed rail', 'Musical styles', 'Bells', 'Fame', 'Search warrants', 'Mayan civilization', 'Political representation', 'Furloughs', 'Stock options', 'Injury prevention', 'Decomposition', 'Workers', 'Electric bicycles', 'Anti-imperialism', 'Economic slowdowns', 'Veterinary colleges', 'Museum exhibits', 'Task forces', 'Fabric analysis', 'Housing prices', 'Convents', 'Blindness', 'Native art', 'Mold', 'Industrialized nations', 'Medical examiners', 'Compact discs', 'Gasoline taxes', 'Palestinians', 'Land economics', 'Cold', 'Ratios', 'Chinese literature', 'Income funds', 'Casting', 'Middle schools', 'Drug prices', 'Law firms', 'Management decisions', 'Syphilis', 'Telegraph service', 'Military policy', 'Self incrimination', 'Baby foods', 'Yield curve', 'Counterinsurgency', 'Cranes & hoists', 'Newspapers', 'Corporate bonds', 'Trout', 'Plumbing fixtures', 'Crop diseases', 'Laicization', 'Emissions control', 'Dividends', 'Fasting', 'Blood banks', 'Medical aid', 'Cigarette industry', 'Gas leaks', 'Infant mortality', 'Neurosurgery', 'Inequality', 'Genetic diversity', 'Larynx', 'Literacy', 'Twin plants', 'Business schools', 'Inflation', 'Masks', 'Judaism', 'Digital imaging', 'Oil sands', 'Military base closures', 'Drought', 'Interactive television', 'Obstetrics', 'Audio equipment', 'Moratoriums', 'Disease', 'Private equity', 'Childrens health', 'Corporate taxes', 'Plankton', 'Print advertising', 'Derivatives', 'Credit reports', 'Business communications', 'Economic reform', 'Arrests', 'Boys clubs', 'Diapers', 'Food service industry', 'Sketches', 'Retirement communities', 'Firefighters', 'Prenatal care', 'Metallurgy', 'Snakes', 'Pharmacology', 'Narrative theme', 'Energy consumption', 'Sunburn & sun tanning', 'Fads', 'Bandwidths', 'Biofuels', 'Atrocities', 'CPAs', 'Coal mining', 'Median', 'Federal Rules of Criminal Procedure', 'Melting', 'Drug abuse', 'Prison libraries', 'Burnout', 'Space telescopes', 'Hiring', 'Megacities', 'Defective products', 'Terminations', 'Limited liability companies', 'Molecular structure', 'Financial services', 'Veganism', 'Art deco', 'Language policy', 'Purchasing contracts', 'Greek civilization', 'Pay for performance', 'Senses', 'Voting Rights Act', 'Scientists', 'Principal components analysis', 'Artificial insemination', 'Egyptology', 'Hmong people', 'Domestication', '16th century', 'Court reporting', 'Child labor', 'Toxins', 'Timber industry', 'Sex discrimination', 'Decks & patios', 'Cyclones', 'Alloys', 'Structured products', 'Distilleries', 'Venue', 'Semantic web', 'Sexuality', 'Connectivity', 'Cruise missiles', 'Steel products', 'Valleys', 'Ideology', 'Factories', 'Dwarfism', 'Medical wastes', 'Age groups', 'Purchase options', 'Epidemics', 'Progressive Era', 'Web analytics', 'Eating disorders', 'Turnover', 'Mercantilism', 'Polyesters', 'Outsider art', 'Shorelines', 'Tuition', 'Registration', 'Sunni Islam', 'Hate speech', 'Bullying', 'Clemency', 'Hydraulics', 'Prisoners', 'Flamenco', 'Mistaken identity', 'Environmental tax', 'Living arrangements', 'Agribusiness', 'Seafood', 'Gallbladder', 'Pain', 'Apheresis', 'Sheep', 'Retailing', 'Blood & organ donations', 'Transportation planning', 'Mortgage rates', 'Vaults', 'Wire transfer', 'Puberty', 'Offshore oil wells', 'Supplies', 'Securities prices', 'Personhood', 'Montage', 'Poets', 'Daughters', 'Year in review', 'Marathons', 'Interoperability', 'War of 1812', 'Arab Israeli relations', 'Beauty salons', 'Displaced persons', 'Idealism', 'Crop science', 'Metropolitan areas', 'Huntingtons disease', 'House brands', 'Robotics', 'Full employment', 'Masonry', 'Pedestrians', 'Euthanasia', 'Physical restraints', 'International lending', 'Entertainment industry', 'Standardization', 'Insurance commissioners', 'Fiduciaries', 'Prescription drugs', 'Bird watching', 'Quantum theory', 'Military history', 'Forced labor', 'Outdoors', 'Agricultural cooperatives', 'Abolitionists', 'Smog', 'Labor costs', 'Agricultural biotechnology', 'Mental disorders', 'Role playing', 'Mumps', 'Farmers markets', 'Venom', 'Risk management', 'Permanent establishment', 'Diabetes', 'Innovations', 'Uric acid', 'Farm buildings', 'Memoranda', 'Family physicians', 'Blood pressure', 'Chief privacy officers', 'Java', 'Concussion', 'Japanese Americans', 'Eating behavior', 'Anaerobic threshold', 'Transfer pricing', 'Workloads', 'Social classes', 'Community', 'Job satisfaction', 'Composers', 'Futures', 'Fencing', 'Mountain biking', 'Brain drain', 'Multimedia', 'Sovereign debt', 'Rhetoric', 'Camcorders', 'Gay rights movement', 'Construction industry', 'Blood transfusions', 'Public libraries', 'Historic preservation', 'Anti-fascism', 'Livestock', 'Fitness training programs', 'Sociopolitical factors', 'Vegetables', 'Voter fraud', 'Social sciences', 'Pet care industry', 'Religious cults', 'Magnesium alloys', 'Economic conditions', 'Waivers', 'Rape', 'Submarine warfare', 'Bipolar disorder', 'Improvisation', 'Professions', 'Gift cards & certificates', 'Older people', 'Screenwriters', 'Programming languages', 'Shells', 'Bridges', 'Virtual communities', 'Zoroastrianism', 'Creeks & streams', 'Student writing', 'Signatures', 'Financial planning', 'Debt service', 'Patent law', 'Antigens', 'Tuberculosis', 'Rapid prototyping', 'Boarding schools', 'Nonresidents', 'University professors', '14th century', 'Television programs', 'Normal trade relations', 'New employees', 'Scooters', 'First ladies', 'Working groups', 'Animal control', 'Native religions', 'Racketeering', 'Military tribunals', 'Government sponsored enterprises', 'Engraving', 'Voyeurism', 'Readers', 'Design specifications', 'Jewelry industry', 'Behavioral economics', 'Carbon fibers', 'Discrimination', 'Eye contact', 'Carbon offsets', 'Sun', 'Memory', 'Digital archives', 'Employee stock ownership plans--ESOP', 'Hot rolling', 'Masonic organizations', 'Horticulture', 'Nonunion', 'Strategic planning', 'Parliamentary committees', 'Dow Jones averages', 'Motion pictures', 'NASDAQ trading', 'Photographs', 'Bus drivers', 'Low income housing credit', 'Soups', 'Cleaning', 'Gymnastics', 'Hearings & confirmations', 'Turkish Americans', 'Graduate students', 'Streaming media', 'Margined securities', 'Economic statistics', 'Digital computers', 'Astrobiology', 'Technology adoption', 'Tellers', 'Journalistic ethics', 'Components industry', 'Spheres', 'Peacekeeping forces', 'Railway networks', 'Body cameras', 'Geology', 'Cowboys', 'Consumer goods', 'Intellectual capital', 'Travelers checks', 'Purchasing power', 'Blood clots', 'Motion detectors', 'Names', 'Social norms', 'Offshore', 'Digitization', 'Data collection', 'Audit departments', '3-D printers', 'Rickets', 'Rainforests', 'Telomerase', 'Multidrug resistant organisms', 'Installment payments', 'Arbitrage', 'Mutual funds', 'Microphones', 'Obesity', 'Interval training', 'Computer simulation', 'Greeting cards', 'Secondary schools', 'Canoes & canoeing', 'Prescribed fire', 'Direct selling', 'Webcasting', 'International conferences', 'Fishing tackle', 'Jockeys', 'Solar eclipses', 'Hydroelectric plants', 'Wage & price controls', 'Investment clubs', 'Speeches', 'White dwarfs', 'Anxieties', 'Fires', 'Legumes', 'Securities buybacks', 'Snow removal', 'Pop art', 'Deities', 'Fiduciary responsibility', 'Gross National Income--GNI', 'Social criticism & satire', 'Marine mammals', 'Consent decrees', 'Service stations', 'Triglycerides', 'Injustice', 'Mortgage insurance', 'Encyclopedias', 'Airports', 'Apologies', 'Carbohydrates', 'International agreements', 'Borrowing', 'Risk sharing', 'Factory outlets', 'Filtering software', 'High income', 'Plant-based beverages', 'Farm loans', 'Black churches', 'Family partnerships', 'Cold storage', 'American Recovery & Reinvestment Act 2009-US', 'Lignin', '21st century music', 'Minimum wage', 'Foreign language instruction', 'No Child Left Behind Act 2001-US', 'Natural resources', 'Rural health care', 'Hard rock music', 'Antidepressants', 'Rheumatoid arthritis', 'Supercomputers', 'Ayurvedic medicine', 'Objectives', 'Self publishing', 'Financial analysis', 'Dolphins & porpoises', 'Chemical elements', 'Violations', 'Religion', 'Wildlife sanctuaries', 'Deferred income taxes', 'Child custody', 'Puzzles', 'Fishing equipment', 'Reptiles & amphibians', 'Predatory lending', 'Blood products', 'Customer services', 'Romantic period', 'Savings & loan holding companies', 'Health care industry', 'Addictions', 'Surrogate mothers', 'Employment interviews', 'Administrative expenses', 'Public relations agencies', 'Synagogues', 'Group insurance', 'Maternity & paternity leaves', 'Digital broadcasting', 'Document management', 'Responsible persons', 'Chlorine', 'Review boards', 'Seven Years War', 'Consumers', 'British Empire', 'Land settlement', 'Poisoning', 'Buses', 'Hip joint', 'Miniature golf', 'Spices', 'Multiple births', 'Rugby', 'Influence', 'Sikhs', 'Aerospace industry', 'Information systems', 'Europeanization', 'Mathematical models', 'Government archives', 'Plastics', 'Space surveillance', 'Prenuptial agreements', 'Economists', 'Higher education', 'Medical research', 'Refrigerators', 'Liberalization', 'Polyphenols', 'Sperm donations', 'Gardens & gardening', 'Small cap investments', 'Affiliates', 'International Financial Reporting Standards', 'Collective action', 'Dining rooms', 'Tequila', 'Intervention', 'Patient-centered care', 'Metabolism', 'Massage', 'Nonrecourse debt', 'Commercial paper', 'Computer to plate technology', 'Humanitarian aid', 'Interdisciplinary aspects', 'Medical screening', 'Birth defects', 'Litigation', 'Firearm accidents & safety', 'Perl', 'Asymmetry', 'Credit default swaps', 'Astrology', 'Food allergies', 'Antibiotics', 'Brutalism', 'Machine translation', 'Grief', 'Cereals', 'Electronic book readers', 'Gaelic football', 'Constructivism', 'Economic indicators', 'Economic analysis', 'Listening comprehension', 'Legal counsel', 'Hearing aids', 'Cosmic rays', 'Missionaries', 'Natural language processing', 'Mayors', 'Fentanyl', 'Communications industry', 'Automobile driving', 'Host country', 'Personality', 'Cleaning compounds', 'Bank Secrecy Act 1970-US', 'VEBA', 'Clarinet music', 'Operating revenue', 'Taboos', 'Civil war', 'Critical thinking', 'Homeopathic medicine', 'Index funds', 'National banks', 'Presidency', 'Corn syrup', 'Focus groups', 'Internet service providers', 'Hypothyroidism', 'Developmental psychology', 'Nobel prizes', 'Mutation', 'Reserve assets', 'Silk', 'Comptrollers', 'Mosques & temples', 'Internet crime', 'Rum', 'Religious right', 'Trade shows', 'Expeditions', 'External debt', 'Econometrics', 'Internet Protocol', 'Endoscopy', 'Equity funds', 'Family owned businesses', 'Effective income tax rates', 'Accidental deaths', 'Succession planning', 'Social networks', 'Power-sharing', 'Birth injuries', 'Fascism', 'Aluminum industry', 'Headgear', 'Fractures', 'School choice', 'Real income', 'Personal shoppers', 'Youth culture', 'Legal defense', 'Built-in gains & losses', 'Skis', 'Price levels', 'Statehood', 'Macular degeneration', 'Middle class', 'Franchisees', 'Traffic violations', 'Ankle', 'Piano', 'Marine biology', 'Mobile commerce', 'Maternal & child health', 'Health services', 'Rupees', 'Pet food', 'Peer to peer lending', 'Bakeries', 'Quinoa', 'Lighthouses', 'Government revenue', 'Farm income', 'Semiconductor research', 'Brand identification', 'Currency', 'Eye protection', 'Jumping rope', 'Sorghum', 'Open access', 'Student teaching', 'Broadband', 'Vehicles', 'Wheelchairs', 'Blood', 'Culture', 'Dark pool trading', 'Merchandising', 'Hydraulic fracturing', 'Federal legislation', 'Mining accidents & safety', 'Tango', 'Spinach', 'Market strategy', 'Foreign policy', 'Marble', 'Public radio', 'Standards', 'Heroism & heroes', 'News wire services', 'Angels', 'Teaching methods', 'Military aircraft', 'Smart cities', 'Lymphoma', 'Community relations', 'Polymers', 'Casualties', 'Mineral rights', 'Search strategies', 'Chief librarians', 'Design', 'Masochism', 'New store openings', 'Footprints', 'Dramatists', 'Advertising pages', 'Disaster relief', 'Humor', 'Precipitation', 'Signs', 'Due process of law', 'Product design', 'Location based services', 'Expanding universe theory', 'Internal auditors', 'Conflict resolution', 'Foreign bank accounts', 'Geographic profiles', 'Algorithms', 'Recapitalization', 'Debt restructuring', 'Projectors', 'Civil aviation', 'Crime prevention', 'Jewelry stores', 'Automated teller machines--ATM', 'Recruitment', 'Clinical medicine', 'Dermatitis', 'Medical technology', 'Domestic violence', 'Workers compensation', 'Hostility', 'Violence', 'Animation', 'Mountain climbing', 'Ventilators', 'Looting', 'Fund raising', 'Foreign operations of US corporations', 'Boarding houses', 'Osteopathic medicine', 'Food waste', 'Soil contamination', 'Libraries', 'Emotions', 'Escrow accounts', 'Assembly lines', 'Boys', 'History education', 'Earthquakes', 'Charitable foundations', 'Film adaptations', 'Artistic movements', 'Global warming', 'Federal Rules of Civil Procedure', 'Tax evasion', 'Cables', 'Meritocracy', 'Mainframes', 'Nuclear reactors', 'Nepotism', 'Cancer surgery', 'Architectural elements', 'Accountability', 'Plea bargaining', 'Selenium', 'Wind power', 'Student teacher relationship', 'Transgender persons', 'Wind farms', 'Retirement homes', 'Roller derby', 'Deodorants', 'Life sciences', 'Corporate raiders', 'Automobile leasing', 'Hostages', 'Principles', 'Tennis rackets', 'Nature', 'Eurozone', 'Library collections', 'Subpoenas', 'Trails', 'Subject fields', 'Countertops', 'Arsenic', 'Human resource management', 'Sufism', 'Assurance services', 'Libel & slander', 'Securities fraud', 'Local government', 'Greenhouse gases', 'Asset backed securities', 'Bomb scares', 'Fisheries', 'Global economy', 'Keywords', 'Music videos', 'Internships', 'Graduate studies', 'Sanitation services', 'Sanctions', 'Automotive parts', 'Progressivism', 'Common law', 'Mental health care', 'Sweating', 'Triathlon', 'Digital video recorders', 'Frequencies', 'Accounting irregularities', 'Biological clocks', 'Exports', 'Social psychology', 'Book binding', 'Photovoltaic cells', 'Falsetto', 'Job openings', 'Fertilizers', 'Religious orthodoxy', 'Martyrs', 'Incentive plans', 'Health care policy', 'Bank management', 'Family leave', 'Theater', 'Online data bases', 'Production increases', 'Intensive care', 'Elevator pitch', 'Unemployment benefits', 'Insanity pleas', 'Syrups & sweeteners', 'Prophecies', 'Bibliotherapy', 'Reflexology', 'Annual reports', 'Elementary schools', 'Discipline', 'Rhythm', 'Opportunity costs', 'Radio communications', 'Microscopy', 'Sadism', 'Energy economics', 'Sporting goods', 'Funding', 'Yield to maturity', 'Fermentation', 'Blogs', 'Pell Grants', 'Motion picture industry', 'Stomach', 'Public good', 'Flexibility', 'English literature', 'Subscribers', 'Electricity distribution', 'Public defenders', 'Generations', 'Mediation', 'Dinosaurs', 'Aircraft carriers', 'Socialism', 'Lava', 'Variable annuities', 'Trade policy', 'Collateralized debt obligations', 'Agnosticism', 'Suicides & suicide attempts', 'Astronomers', 'Iranian literature', 'Guidebooks', 'Medicine', 'District attorneys', 'Checking accounts', 'Recovery (Medical)', 'Airway management', 'Emerging markets', 'Airplane racing', 'Readership', 'Bible', 'Emergency procedures', 'Firewalls', 'Secession', 'Public lands', 'Surrealism', 'Agreements', 'Firearms', 'Archaeology', 'Asian history', 'Lawns', 'Tax exempt organizations', 'Group dynamics', 'Minors', 'Sitar music', 'Presidential elections', 'Harvest', 'Meat products', 'Charitable tax deductions', 'Internet access', 'Benchmarks', 'Nuclear energy', 'Yoga', 'Peace negotiations', 'Computer forensics', 'Research & development--R&D', 'Video equipment', 'Councils', 'Career preparation', 'Gold mines & mining', 'Anti dumping tariffs', 'Hedge funds', 'Shoes & boots', 'Augmented reality', 'Inclusive education', 'Janitors', 'Expenditures', 'Electric rates', 'Pay structure', 'Personal income', 'Expressionism', 'Rabbits', 'Carriages', 'Hydroelectric power', 'Net losses', 'Working class', 'Furnaces', 'Liquor industry', 'Adipocytes', 'State of the Union Address', 'Small mammals', 'Distribution costs', 'Margarine', 'Personal health', 'Ammonia', 'Embezzlement', 'Computers', 'Central business districts', 'Jurisdiction', 'Glass substrates', 'Artillery', 'Genetic disorders', 'Financial inclusion', 'Order processing', 'Petroleum marketers', 'Big Bang theory', 'Cold remedies', 'X rays', 'Luxuries', 'Mental depression', 'Executive compensation', 'Sugarcane', 'Fossils', 'Watergate affair', 'Theocracy', 'Upper class', 'Oil service industry', 'Corporate headquarters', 'Hazardous substances', 'Outside directors', 'Public hearings', 'Stock market delistings', 'Lead content', 'Psychosis', 'Middle age', 'Signal processing', 'Dormitories', 'Placebo effect', 'Offshore drilling', 'Germplasm', 'Treasury bonds', 'Organizational behavior', 'Public property', 'Primaries & caucuses', 'Nonperforming loans', 'Cash basis accounting', 'Business costs', 'Helmets', 'Testimony', 'Alcoholism', 'Opera houses', 'Contingent fees', 'Celebrities', 'Sick leave', 'Tourism', 'Military personnel', 'Home loans', 'Product reliability', 'Business intelligence', 'Bodybuilding', 'Security management', 'Registration statements', 'Artificial intelligence', 'Political crimes', 'Sports fans', 'Crude oil prices', 'Body art', 'Ocean traffic', 'Online entertainment', 'Gases', 'Judaic studies', 'Public relations', 'Drug trafficking', 'Parades', 'Queens', 'Social security numbers', 'Surplus government property', 'Phenology', 'Military readiness', 'One-act plays', 'Moral hazard', 'Feminism', 'Employee leasing', 'Applied arts', 'Skepticism', 'Reinsurance', 'Dental implants', 'Girls education', 'Witness protection programs', 'Laxatives', 'Feet', 'Tsunamis', 'Androgens', 'Shareholders rights', '12th century', 'Economic activity', 'Technological change', 'Criminal liability', 'Hanukkah', 'Authenticity', 'Government grants', 'Communism', 'Pro life movement', 'Legal arguments', 'Serial murders', 'Military benefits', 'Cross selling', 'Islamic financing', 'Civil unions', 'Partisanship', 'Color', 'Black Lives Matter movement', 'Convertible debentures', 'Class size', 'Hospitality industry', 'Psychics', 'Elective surgery', 'Kosher food', 'Religious movements', 'Grasslands', 'Cold War', 'Toy industry', 'Emergency services', 'Silicon carbide', 'Batteries', 'Radiation therapy', 'Surface decoration', 'Sewing machines', 'Political economy', 'Wall Street Reform & Consumer Protection Act 2010-US', 'Rehabilitation of criminals', 'Renminbi', 'Defense industry', 'Reserve requirements', 'Kickbacks', 'Coins', 'Prisoner treatment', 'Sprinkler systems', 'Kennels', 'Rule of law', 'Guards', 'Militancy', 'Wealth tax', 'Area codes', 'Pharmacy benefit management', 'Welfare reform', 'Fibromyalgia', 'Chinese culture', 'Emission standards', 'Dietary guidelines', 'Television ratings', 'Animal behavior', 'Ports', 'Business expenses', 'Minimalism', 'Tropical diseases', 'Crops', 'Federalism', 'Visual arts', 'Exploitation', 'JavaScript', 'Marine conservation', 'Responsibilities', 'Live performance', 'Agricultural research', 'Communitarianism', 'Water transportation', 'Inflammatory bowel disease', 'Characters (Roles)', 'Collectivism', 'Military recruitment', 'Vampires', 'Identification', 'Floor coverings', 'Low income groups', 'Recreational equipment', 'Silicon', 'Science education', 'Collateralized bond obligations', 'Foundations', 'Bows & arrows', 'Corporate tax planning', 'Coal', 'Accounting', 'Municipal bonds', 'General aviation', 'Herbal medicine', 'Cargos', 'Wearable computers', 'Trees', 'Camping', 'Labor standards', 'Deflation', 'Etiquette', 'Interactive marketing', 'Air flow', 'Christian Islamic relations', 'Alliances', 'Buildings', 'Painters', 'Fair trade', 'County ordinances', 'Double taxation', 'Petroleum industry', 'Men', 'Asset allocation', 'Deep learning', 'Decentralization', 'Three dimensional imaging', 'Credit policy', 'Pornography & obscenity', 'Performance appraisal', 'Copyright', 'Gerontology', 'Underground Railroad', 'Loyalty programs', 'Access to materials', 'Hyperhidrosis', 'Toothbrushes', 'Economic trends', 'Philosophers', 'Primary care', 'Access to education', 'Auditing standards', '17th century', 'Shinto', 'Joint replacement surgery', 'Securities industry', 'Government spending', 'Detectives', 'Drug use', 'Drivers licenses', 'Population density', 'Retirement policies', 'Profit sharing plans', 'Vending machines', 'Self evaluation', 'Fast food', 'Finished goods', 'Revisions', 'Body temperature', 'Laparoscopy', 'Country clubs', 'Solvency', 'Newspaper industry', 'Trucking industry', 'Automobile racing', 'Male employees', 'Baked goods', 'Family structure', 'Response time', 'Chefs', 'Cotton', 'Sculpture', 'Stainless steel', 'Infertility', 'Drunk driving', 'BSE', 'Motivation', 'Automobile clubs', 'Grasses', 'Art dealers', 'Immersion programs', 'Ecology', 'Foreign residents', 'Native species', 'Nurses', 'Parental rights', 'Wool industry', 'Lyrics', 'Whites', 'Shillings', 'Trade surplus', 'Patient safety', 'Clerical personnel', 'Food irradiation', 'Construction loans', 'Dairy products', 'Licensing examinations', 'Field representatives', 'Mounted police', 'Municipal employees', 'Capital losses', 'Neptune', 'Documentary films', 'Victorian period', 'Genealogy', 'Alternative energy', 'Yiddish language', 'Saws', 'Unsolved crimes', 'Working hours', 'Blasphemy', 'Art history', 'Industrial research', 'Public buildings', 'Biomass energy', 'Educational leadership', 'Human rights', 'Technology stocks', 'Inventory management', 'Corporate presidents', 'Recessions', 'Suicide bombings', 'Plumbing', 'Absolute pitch', 'Defects', 'Ball bearings', 'Cellular telephones', 'Manuscripts', 'Cosmetology', 'Title insurance', 'Middle Ages', 'Divestiture', 'Value stocks', 'Body image', 'Stone & clay industries', 'Lute music', "Coups d'etat", 'Adaptation', 'Pulp & paper industry', 'Poverty', 'Business analytics', 'Congenital diseases', 'Zoos', 'Legal fees', 'Medical tests', 'Rates of return', 'Organizational change', 'Posters', 'Court records', 'Public interest', 'Roofing', 'Logic', 'Bulimia', 'Quantitative analysis', 'Weather forecasting', 'Environmental management', 'Multiplication & division', 'Sabbatical leave', 'Public television', 'Acid rain', 'Typhoons', 'Injuries', 'Tattoos', 'Industrial economics', 'Birth weight', 'Dog racing', 'Working conditions', 'Admissions policies', 'Deaths', 'Medicare', 'Extinction', 'Surveillance', 'Fingerprinting', 'Benefit cost analysis', 'Cardiac stress tests', 'Farms', 'Ostracism', 'Rockets', 'Tax refunds', 'Big Data', 'High rise buildings', 'Racial profiling', 'Lobbying', 'EU membership', 'Bank examiners', 'Shadow banking', 'Commercial markets', 'Sale of a business', 'Elevators & escalators', 'Sex roles', 'Plaster', 'Currency transactions', 'Juvenile offenders', 'Funding cuts', 'Federal funds rate', 'Trauma', 'Russian Revolution', 'Vitamin A', 'Plein air', 'Rice', 'Vasectomy', 'Judicial appointments', 'Ethnicity', 'Teeth', 'Eroticism', 'Electoral reform', 'Fire stations', 'Equity stake', 'Dredging', 'Blacklisting', 'Restraining orders', 'Embedded systems', 'Clocks & watches', 'Wrestling', 'Behavior', 'Fund accounting procedures', 'Vernal equinox', 'Use taxes', 'Co-branding', 'Hematoma', 'Military awards', 'Methods', 'Ghettos', 'Shareholder relations', 'Art criticism', 'Forbearance', 'Pilots', 'Leaves of absence', 'Weapons of mass destruction', 'Price cuts', 'Chairs', 'Medical practices', 'Real estate appraisal', 'Infectious diseases', '3-D technology', 'Genomes', 'Streptococcus infections', 'Enhanced oil recovery', 'Cemeteries', 'Electromagnetism', 'Members of Parliament', 'Asthma', 'Feedback', 'High strength steel', 'Vitamins', 'Glaucoma', 'Calculus', 'Menopause', 'Taxes', 'Foreign partnerships', 'Criminals', 'Office management', 'Thoracic surgery', 'Climate science', 'Crystallography', 'Gonorrhea', 'Wines', 'Boxes', 'Post-traumatic growth', 'Malnutrition', 'Assessed valuation', 'Field hockey', 'Oligarchy', 'Environmental health', 'Marimba music', 'Composting', 'Consolidated financial statements', 'Sin', 'Relativism', 'Drug resistance', 'Scripts', 'Underwater pipelines', 'Ground stations', 'Etching', 'Regulated industries', 'Antiquarian materials', 'Cassava', 'Political conventions', 'User behavior', 'Habits', 'Musical instruments', 'Banking law', 'Public health', 'Budgeting', 'Naturalism', 'Savings bonds', 'Heroin', 'Evolutionary biology', 'Osteoarthritis', 'Digital technology', 'Porcelain', 'Southeast Asian studies', 'Insurance claims', 'Cryptography', 'Middle Eastern studies', 'Right to counsel', 'Seeds', 'Weapons', 'Chemical plants', 'Early retirement', 'Initial public offerings', 'Literary history', 'Whales & whaling', 'Pro bono services', 'Foster care', 'Stalking', 'Cultural property', 'Meteors & meteorites', 'Provenance', 'Telecommuting', 'Labor contracts', 'Emigration', 'Summons', 'Energy conservation', 'Fire prevention', 'Diamond industry', 'Merit increases', 'Business etiquette', 'Basins', 'Professional basketball', 'Shipments', 'Confessions', 'Bond issues', 'International relations', 'Maple syrup', 'Educational software', 'Transmitters', 'Truces & cease fires', 'Leisure', 'Highway construction', 'Student financial aid', 'Health care access', 'Drugs', 'Common markets', 'Name changes', 'Underage drinking', 'Gangrene', 'Light', 'Prisoners of war & missing in action', 'Multiple sclerosis', 'Enterprise resource planning', 'Womens rights', 'Biomass', 'Quantum computing', 'Immunoglobulins', 'Heat', 'Carpentry', 'Financial statements', 'Wives', 'Seawater', 'Cookbooks', 'Smart grid technology', 'Intelligence gathering', 'Competition policy', 'Global depositary receipts', 'Secret police', 'Government subsidies', 'Options trading', 'Piano music', 'Interstate Commerce Act-US', 'White collar workers', 'University presidents', 'Games', 'Student housing', 'Provincial government', 'Political dissent', 'Business information', 'Economic growth', 'Tax credits', 'Colds', 'Arbitration', 'Casinos', 'Lending', 'Parent teacher groups', 'Electronic health records', 'Interactive computer systems', 'Private investigators', 'Multiemployer pension plans', 'Gold', 'Product tampering', 'Student productions', 'Hyperinflation', 'Fire departments', 'Media literacy', 'Financial institutions', 'Liturgy', 'Assistance animals', 'Criminal records', 'Picnics', 'Papal visits', 'Tariff agreements', 'Wages & salaries', 'Management styles', 'Administrative reform', 'Discovery rule', 'Tomatoes', 'Commissioned works', 'Federal budget', 'News media', 'Mushrooms', 'Hypotheses', 'School admissions', 'Landfill', 'Aerospace medicine', 'Mock conventions & elections', 'General partners', 'Hydrogen', 'Loneliness', 'Shoreline protection', 'Salmonella', 'Conferences', 'Release dates', 'Humanism', 'Orientalism', 'Bears', 'Coronaviruses', 'Sermons', 'Inventory control', 'Paratext', 'Zoning', 'British people', 'Down payments', 'Hearing impairment', 'Haute couture', 'Spacecraft', 'Social workers', 'Geoengineering', 'Bankers associations', 'Case law', 'Social costs', 'Emojis', 'Boxing', 'Algae', 'Breast cancer', 'Klezmer music', 'Writing', 'Productivity', 'Immunosuppressive agents', 'Guardians', 'Romances', 'Historic documents', 'Payment systems', 'Advertising', 'Tumors', 'Experiential learning', 'Homeland Security Act 2002-US', 'Expected utility', 'Blood cancer', 'Book industry', 'Fungi', 'Collagen', 'Cartography', 'Puritanism', 'Cheese industry', 'Risk premiums', 'Parks & recreation areas', 'Aquariums', 'Personality traits', 'Cement industry', 'Global funds', 'Cyanobacteria', 'Fares', 'Stock market indexes', 'Railroad crossings', 'Nonsteroidal anti-inflammatory drugs', 'Homeowners', 'American dollar', 'Amputation', 'Medical prognosis', 'Bookstores', 'Intellectuals', 'Collusion', 'Pearls', 'Specialization', 'Hair loss', 'Industrial Revolution', 'Bones', 'Fast food industry', 'Solitary confinement', 'Oral history', 'Pensions', 'Decriminalization', 'Martial arts', 'Railroad transportation', 'Comets', 'Beds', 'Existentialism', 'Americans abroad', 'Assaults', 'Legislatures', 'Custom design', 'Middle management', 'College professors', 'Bank ratings', 'Academic degrees', 'Native peoples', 'Prairies', 'Prescriptions', 'Reproductive sterilization', 'Relief provisions', 'Cartilage', 'Magnetism', 'Constitutions', 'World history', 'Web 2.0', 'Skating', 'Six Sigma', 'Low density lipoprotein', 'Employment security', 'Political advertising', 'Estuaries', 'Hospitalization', 'Initiatives', 'Umbilical cord', 'Space debris', 'Growth stocks', 'Golden Globe awards', 'Fat substitutes', 'Racquetball & squash', 'Sectarian violence', 'Trade adjustment assistance', 'Roadblocks', 'Brain diseases', 'Housing', 'Buttons', 'Citizen participation', 'Diplomatic & consular services', 'Butterflies & moths', 'Genetically altered foods', 'Hands', 'Operating systems', 'Protectionism', 'Acting', 'Softball', 'Photographers', 'Research & development expenditures', 'Mathematics', 'Theology', 'Cashews', 'Insecticides', 'Money markets', 'Works councils', 'School dropout programs', 'Dialectics', 'Archivists', 'Museums', 'Computer security', 'Book awards', 'Suspicious activity reports', 'Nurse practitioners', 'Treasury bills', 'Cooking', 'Tennis equipment', 'Academic libraries', 'Retained earnings', 'Lamps', 'Public life', 'Parole & probation', 'Military deployment', 'Governors', 'Alternative medicine', 'Best practice', 'Cash flow statements', 'Administrative law', 'Hoarding disorder', 'Disorderly conduct', 'Sensory perception', 'Medical errors', 'Hedging', 'Online advertising', 'Single parents', 'Genetics', 'Contests', 'Religious music', 'False arrests & convictions', 'Institutionalization', 'Maritime law', 'Consciousness', 'Multilingualism', 'Modernism', 'Irritable bowel syndrome', 'Personal grooming', 'Particle physics', 'Brain cancer', 'Operating leverage', 'Legal reform', 'Rocks', 'Union leadership', 'Container ships', 'Assisted suicide', 'Equity participations', 'Physical fitness', 'Eastern Orthodox churches', 'Telecommunications industry', 'Ice cream', 'Childrens television', 'Legionnaires disease', 'Race', 'Textbooks', 'Long term health care', 'Beans', 'Homeowners insurance', 'Criminal statistics', 'Boats', 'Pioneers', 'Vikings', 'Sexually transmitted diseases--STD', 'Socks', 'Surgery', 'Survivalists', 'Phonograph records', 'Agricultural commodities', 'Lactose', 'Effigies', 'Sea level', 'Eye surgery', 'Public assistance programs', 'Child pornography', 'Vacation homes', 'Reforms', 'Minority & ethnic violence', 'Metal forming', 'Typing', 'Canon law', 'Pollution', 'Compressed natural gas', 'Microfinance', 'International comparisons', 'Braking systems', 'Rodeos', 'Treasury operations', 'Captive finance', 'Amyotrophic lateral sclerosis', 'Basic income', 'Automobile industry', 'Elizabethan period', 'LNG', 'Price ceilings', 'Civil liberties', 'Cost reduction', 'Law clerks', 'Put & call options', 'Hypoglycemia', 'Letters of intent', 'Corporate officers', 'Environmental science', 'Business community', 'Human resources', 'American literature', 'Poll taxes', 'Balance', 'Estrogens', 'Pest control', 'Recycling', 'Genre', 'Housing starts', 'Hoaxes', 'Baroque era', 'Runoff', 'Retirement planning', 'Korean War', 'Parkinsons disease', 'Dolls', 'Occupational mobility', 'Clouds', 'Criminal investigations', 'Glaciers', 'Office software', 'Acoustics', 'Freedom of speech', 'Palimony', 'Cardiomyopathy', 'Synthetic products', 'Telephone service', 'School busing', 'Property taxes', 'Dollar standard', 'Language teachers', 'Basketball', 'Digital curation', 'Mollusks', 'Mandolin music', 'Dancers & choreographers', 'Military schools', 'Government agencies', 'Postmodernism', 'Modernity', 'Autistic children', 'Wool', 'Advertising rates', 'Confidentiality', 'Consumer organizations', 'Corporate reorganization', 'Sports arbitration', 'Panoramas', 'Wood products', 'Prototypes', 'Motor ability', 'Literary characters', 'Farm price supports', 'Cocaine', 'Gift shops', 'Tenants', 'Bamboo', 'Radio broadcasting', 'Foreign Account Tax Compliance Act 2010-US', 'Seasonal variations', 'Endorphins', 'Part time employment', 'Coronary vessels', 'Community development', 'Cockroaches', 'Smartphones', 'Net present value', 'Indian culture', 'Standardized tests', 'Herbicides', 'Body fat', 'Violent crime', 'International finance', 'Apartheid', 'Standard of living', 'Animal communication', 'College football', 'Professional football', 'Ghostwriting', 'Weather', 'Prepaid services', 'Defined contribution plans', 'Nuclear accidents & safety', 'High occupancy vehicle lanes', 'Affluence', 'Private placement', 'Diaspora', 'Inflammation', 'Industrial policy', 'Lanham Act 1946-US', 'Corporate sponsorship', 'Disease prevention', 'Libertarianism', 'Fetal alcohol syndrome', 'SEC regulations', 'Image', 'Nation states', 'Microwaves', 'Investment', 'Bar associations', 'Pore size', 'Sustainable development', 'Copper industry', 'Regulatory reform', 'Land reclamation', 'Value added', 'Vitamin D', 'Ventilation', 'Presidential powers', 'Risk aversion', 'Amphitheaters', 'Pumpkins', 'Employee savings plans', 'Wilderness areas', 'Terms of trade', 'Collateral', 'Population decline', 'Saunas & hot tubs', 'Performance art', 'Management consultants', 'Sales taxes', 'Fair presentation', 'Arts and crafts movement', 'Venus', 'Transplants & implants', 'Aerosols', 'Drug legalization', 'Infections', 'Fire hazards', 'Reconciliation', 'Passenger screening', 'Underground storage', 'Steel alloys', 'Gold standard', 'Chemotherapy', 'Psychology', 'Medical laboratories', 'English language', 'Translations', 'Engineering firms', 'Biofeedback', 'Diodes', 'Appointments & personnel changes', 'Extraterrestrial life', 'Sleep deprivation', 'Agricultural subsidies', 'Hospital ships', 'Food prices', 'Athletes', 'Salespeople', 'Pheromones', 'Lung diseases', 'Vehicle transmissions', 'Spirituality', 'Finance companies', 'Biblical studies', 'Auctions', 'Anniversaries', 'Bullion', 'Roma', 'Bank failures', 'Military helicopters', 'Social interaction', 'Circumcision', 'Jews', 'Chemical compounds', 'Market erosion', 'Life expectancy', 'Environmental stewardship', 'Offset printing', 'Truck drivers', 'Aging', 'Cognitive ability', 'Organic light emitting diodes', 'Horse racing', 'Immunotherapy', 'Social research', 'Carp', 'Gamification', 'Ecumenism', 'Wrap accounts', 'Construction accidents & safety', 'Reality programming', 'Community banks', 'Minority stockholders', 'Fair use', 'Millennials', 'LIBOR', 'Lipids', 'Cantatas', 'Failure', 'Handbooks', 'Recall of government officials', 'Socioeconomic factors', 'Annuities', 'Medical malpractice', 'American culture', 'Market economies', 'Leukemia', 'Criminal procedure', 'Homophobia', 'Virulence', 'Exercise', 'Gentrification', 'Congressional committees', 'Autoimmune diseases', 'Electric power lines', 'Dance', 'Gospel music', 'Gum disease', 'Handwriting', 'Condominiums', 'Environmental cleanup', 'Reconstructive surgery', 'Motor vehicle fleets', 'Sulfur content', 'International relations-US', 'Riots', 'Evacuations & rescues', 'Plant diseases', 'Social change', 'Female circumcision', 'Older workers', 'Halo effect', 'Roles', 'Self image', 'Prejudice', 'Payroll costs', 'Information management', 'Protein folding', 'Trade disputes', 'Bankruptcy claims', 'Digital asset management', 'Patients rights', 'Visitor centers', 'Variable interest rates', 'Xenophobia', 'Product development', 'Grain', 'Sovereignty', 'Political behavior', 'Harpsichord music', 'Muscle fatigue', 'Skin cancer', 'Genetic testing', 'Labor supply', 'GST', 'High frequency trading', 'Gallium arsenide', 'Hydrogels', 'Blood groups', 'Central banks', 'Fractals', 'Business metrics', 'Arab Americans', 'Classical music', 'Injunctions', 'Overseas employment', 'Longitudinal studies', 'Illegal drug operations', 'Upward mobility', 'Heatstroke'}
In [53]:
## from BKMX paper 
subject_labels = pd.read_csv('filters/djn_subject_labels.csv',
                                 header=None)[1].to_list()

subject_labels
Out[53]:
['description',
 'Biography',
 'Basketball',
 'Baseball',
 'Colleges & Universities',
 'Farnborough Air Show',
 'Fashion',
 'Football',
 'Golf',
 'Getting Personal',
 'Hockey',
 'Letters to the Editor',
 'Lifestyles',
 'Obituaries',
 'Religion',
 'Soccer',
 'Sports',
 'Sports & Recreation',
 'Tabular Material',
 'Test Message',
 'Tennis',
 'Travel']
In [54]:
## subject filter first time 

subject_list = ['Sports',
                'Sport',
                'Game',
                'Olympics',
                'Olympic',
                'Leisure',
                'Weekends',
                'Yacht',
                'Mansion',
                'Resorts & spas',
                'Cooking',
               'Food',
                'Wine',
               'Travel',
               'Book',
                'Books',
                'Literature',
                'Love',
                'Fiction',
                'Writer',
                'Novel',
                'Arts',
                'Art',
                'Musuem',
                'Restaurant',
                'Film',
                'Opera',
                'Theater',
                'Music',
                'TV',
               'Obituary',
               'Poetry',
               'Culture',
                'Religion',
                'Dance',
                'Exhibition',
                'Culture commentary',
               'Book reviews',
               'Basketball',
               'Boxing',
               'NBA',
               'Fashion',
               'Poetry',
                'Philosophy',
               'NFL',
               'Tennis',
                'Soccer',
               'Style',
               'Marriage',
               'Abortion',
               'Phisics',
               'Psychology',
               'History',
               'Justice',
               'Police',
               'FIFA',
               'Crime',
               'Laws',
               'Constitution',
               'Court',
               'Supreme court',
               'Shooting',
               'Guns',
                'Gun',
               'Race',
               'Gender',
               'Children',
               'Education']

subject_list = subject_list + subject_labels

for subject in subject_list:
    article_data = exclude_subjects(subject,
                                    article_data)
In [55]:
len(article_data)
Out[55]:
84349
In [56]:
## headline patterns filter

headlines_patterns = pd.read_csv('filters/regex_drop_headlines.csv',
                                 header=None)[0].to_list()

for pattern in headlines_patterns:
    print(pattern)
    article_data = exclude_headlines(pattern,
                                    article_data)
ad note
adnotes
architecture:
architecture -
arena:
arena -
arts & enter
art:
art -
asides:
asides -
author q&a:
author q&a -
auto sales data
automobile sales statistics
bay area:
bay area -
best of the law blog
best on the street
best sellers
best selling books
bestsellers
book excerpt:
book excerpt -
bookmarks
books:
books -
book publishing:
bookshelf
box office
broadway:
broadway -
business bookshelf
business educ
career journal
careers:
careers -
cd yields
changes in stockhold
\(chart\)
city news:
city news -
closed end fund
closed-end fund
comment:
comment -
conflicting claims:
conflicting claims -
correction
country doctor:
country doctor -
credit ratings
critique:
critique -
cubicle culture
culture:
dance:
dance -
design:
design -
died\.\.\.
dividend news:
dividend news -
earnings surprises
earnings watch
editor's note
education:
education -
entertainment:
entertainment -
fashion journal:
fashion journal -
fashion:
fashion -
film:
film -
fiscally fit:
food & drink
food:
food -
form \+ function
friday journal
gallery:
gallery -
gardening
getting started:
global finance \(a special report\)
golf:
golf -
greater new york watch
health & wellness:
health & wellness -
health journal
heard & scene
holiday notice
home & family:
home & family -
how poll was conducted
how polls were conducted
in concert
inside baseball:
inside baseball -
insider trading spot
ipo scorecard
jazz:
jazz -
key interest rates
law journal:
law journal -
law brief:
law brief -
leisure & art
leisure brief
leisure:
leisure -
letters to the editor
life & arts
lipper fund indexes
lipper indexes
long toss:
long toss -
managing motherhood
managing your career
mansion:
mansion -
media:
media -
medicine & health
mobile guide:
mobile guide -
money rates
movies:
movies -
music:
music -
mutual fund scorecard
mutual-fund scorecard
new movies
new securities
new stock list
no headline available
notable & quotable
noted\.\.\.
notice to reader
nyse high
obituar
off duty:
off duty -
off the charts
olympics
on disc:
on disc -
on disk:
on disk -
on records:
on records -
on sports
on video:
on video -
open house
opera:
opera -
on opinionjournal.com
opinionjournal.com
people pattern
pepper and salt
pepper\.\.\.
photography:
photography -
private properties
program trading
property monday
property watch
property:
property -
publicly traded
publishing:
publishing -
quarterly earnings surprises
recent changes in stock
recent sec
recent stock-list
redemption notic
review & outlook \(editorial\):
review & outlook\(editorial\):
review outlook
review /
review / recordings:
review / recordings -
review:
review -
rule of law:
rule of law -
science journal
securities offering calendar
shop talk:
shop talk -
shoptalk:
short interest high
spaces:
spaces -
sports
stock ex-div
stocks ex-div
style & sub
style & travel:
style & travel -
style:
style -
summer olympics
takeoffs & land
taste:
taste -
tastings:
tastings -
tax-exempts
television:
television -
the buzz:
the buzz -
the daily fix
the gallery:
the gallery -
the informed reader
the mobile guide
the most popular stories on the web
the property report
the ticker
the week ahead
the weekend int
theater
theatre
time off / history:
time off / history -
time off:
time off -
timeoff:
today's agenda
top ten singles
top ten albums
tracking travel
travel:
travel -
treasury auction
tv preview:
tv preview -
tv:
tv -
u\.s\. market to close
watching the web
weekend invest
weekend journal
weekend report
what's news:
what's news -
what's your workout
who's news
wine:
wine -
winter olym
work & family
workspaces
In [57]:
len(article_data)
Out[57]:
80404
In [58]:
## after filtering 
article_data.sample(n=30)
Out[58]:
Article ID text date subjects title month_date day_of_week length
69311 1353214806.xml AFGHANISTAN Suicide-Bomb Attack Kills P... 2013-05-21 Suicide bombings,Fatalities,Councils,Poliomyel... World News: World Watch 2013-05 1 4051
26716 1691887792.xml Minsk, Belarus -- Belarusian geologist ... 2015-06-30 Stone,Councils,Rocks,Museums, Off the Wall: Belarus Museum Is Between Rock A... 2015-06 1 6294
55616 308819086.xml MEXICO CITY -- Mexican police have arres... 2010-05-28 Cartels,Organized crime,Airports,Criminal inve... World News: Mexico arrests mayor of Cancun 2010-05 4 2729
93680 1535868589.xml VEVEY, Switzerland -- Nestle SA wants t... 2014-06-17 Marketing,Consumers,Coffee industry, Nestle Aims to Perk Up Nescafe 2014-06 1 4232
9154 308819737.xml Britain's new government on Thursday pro... 2010-05-28 Unemployment benefits,Gross Domestic Product--... Europe News: Welfare will be target of U.K. be... 2010-05 4 5961
45467 1550895714.xml New Delhi -- Mahesh Nath is a governmen... 2014-08-05 Members of Parliament,Monkeys & apes,Satellite... Off the Wall: 'Monkey Wallahs' Around New Delh... 2014-08 1 5545
63441 756164127.xml BRUNETE, Spain -- This leafy town on the... 2010-10-04 Economic development,Sovereign debt,Employees,... Europe News: Spanish towns struggle under debts 2010-10 0 7098
26177 1039300085.xml SHANGHAI -- Italian auto manufacturer F... 2012-09-14 Production capacity,Automobiles, Fiat Launches New Compact in China 2012-09 4 3978
82316 1727796814.xml WASHINGTON -- Republicans nominated Rep... 2015-10-29 Budgets,Primaries & caucuses,Nominations,Elect... U.S. News: Ryan Gets GOP Nod For House Speaker 2015-10 3 3294
24648 912069871.xml LONDON -- CNN talk-show host Piers Morga... 2011-12-21 Hackers,Scandals, Piers Morgan Grilled at U.K. Inquiry 2011-12 2 3921
4662 1881331105.xml Depending on how things go in the April... 2017-03-28 Presidents,Dictators,Elections, Americas: Julian Assange and Ecuador's Election 2017-03 1 4842
65435 1473677508.xml Glendale, Calif. -- At just 8 years old... 2014-01-03 Social networks,Fund raising,Bears,New year,Pi... Off the Wall: Spat Over a Trashy Celeb's Fame ... 2014-01 4 6247
80400 1545119912.xml A broad majority of pro-European Union ... 2014-07-16 Prime ministers,Legislators,Political parties, Europe News: Plethora of Promises Help Juncker... 2014-07 2 3371
85266 1787010962.xml Generic drugs have long delivered huge ... 2016-05-06 Pharmaceutical industry,Sales,Brand names,Cost... Biotech Drug Knockoffs Bring Paltry Savings --... 2016-05 4 9982
80368 308826972.xml The forced march to pass ObamaCare conti... 2010-03-18 Tax rates,Excise taxes,Economic development,Pa... ObamaCare's Worst Tax Hike 2010-03 3 4206
22768 1336969127.xml CONSUMER GOODS Clark Takes SABMiller He... 2013-04-24 Corporate profits,Stockholders,Shareholder mee... Business Briefs 2013-04 2 6297
55358 921684494.xml Since MF Global Holdings Ltd. filed for ... 2012-02-17 Bankruptcy,Customers, MF Global Claims Are Luring Investors 2012-02 4 3679
74413 1790294592.xml TEL AVIV -- Israel's prime minister, Be... 2016-05-23 Energy industry,National security,Prime minist... Business News: Israel Spurs Natural-Gas Efforts 2016-05 0 2978
90872 1728164318.xml Visa Inc. is close to clinching the big... 2015-10-30 Visa Closes In on $22 Billion Deal --- Payment... 2015-10 4 2756
24198 1431601811.xml Pakistan said it would release the Afgh... 2013-09-12 Leadership,Peace negotiations,Insurgency,Intel... World News: Pakistan Agrees to Free Senior Tal... 2013-09 3 5675
83141 811820871.xml North Korea's unprovoked attack Tuesday ... 2010-11-26 Stock exchanges,Free trade,Trade agreements,Ta... South Korea's Prosperity Defense 2010-11 4 3920
46297 1738172610.xml POLITICS Ryan Urges Revamp of Mental-He... 2015-12-02 State court decisions, U.S. News: U.S. Watch 2015-12 2 1604
78807 1813833804.xml Rovio Entertainment Ltd., the Finnish m... 2016-08-25 Earnings,Computer & video games,Remakes & sequ... Rovio Sets Sequel To 'Birds' Movie 2016-08 3 1498
48795 1171257709.xml Limiting personal income-tax deductions... 2012-11-20 Taxes,Tax rates,Donations,High income,Tax cuts... U.S. News: An Uneven Bite From Fewer Tax Breaks 2012-11 1 5139
16847 1514684060.xml \n\n\n \n\n\n\nTop U.S. and Japanese trade neg... 2014-04-11 World-Wide 2014-04 4 1256
80386 1444695746.xml For years, Julie Brinton's days have be... 2013-10-25 Motivation,Couples, Bonds: On Relationships: Why We Have Sex (Beyo... 2013-10 4 6385
70469 1772037563.xml ASSICURAZIONI GENERALI Donnet Is in Lin... 2016-03-11 Options markets, Finance Watch 2016-03 4 1539
37673 1652594541.xml Same-sex marriages began in Alabama Mon... 2015-02-10 Licenses,Attorneys general,Same sex marriage,S... U.S. News: High Court Allows Gay Marriages In ... 2015-02 1 2761
80318 1289216857.xml A week from now, a dramatic new U.S. po... 2013-02-21 Budgets,Tax revenues,Tax reform,Committees,Pre... Failed U.S. Leadership Created the Budget Crisis 2013-02 3 5438
69142 1755402840.xml BRUSSELS -- European Union regulators h... 2016-01-11 Antitrust,US exports,Electronic commerce,Econo... FedEx-TNT Express Deal Clears Hurdle 2016-01 0 3965

2. Data Preprocessing

  1. Replace all non-alphabetical characters with an empty string and set the remaining characters to lower-case.
  2. Parse article text into a white-space-separated word list retaining the article’s word ordering. Exclude single-letter words.
  3. Concatenate articles with the same accession-number as these are chained articles.
  4. Remove common “stop” words and URL-based terms. List of exclusions is standard but available from authors on request.
  5. Lastly, we conduct light lemmatizing of derivative words. The following rules are applied in the order given, where ’x’ is a candidate term. In each case, the stemming is only applied if the multiple terms reduce to the same stem.
    • (a) Replace trailing “sses” with “ss”
    • (b) Replace trailing “ies” with “y”
    • (c) Remove trailing “s”
    • (d) Remove trailing “ly”
    • (e) Remove trailing “ed.” Replace remaining trailing “ed” with “e”
    • (f) Replace trailing “ing” with “e”. For remaining trailing “ing” that follow a pair of identical consonants, remove “ing” and one consonant. Remove remaining trailing “ing”
    • (g) Remove words with less than 3 letters.
  6. From the resulting uni-grams, generate the set of bi-grams as all pairs of (ordered) adjacent uni-grams.
  7. Exclude terms (both uni-grams and bi-grams) appearing in less than 0.1% of articles. The unique set of terms is the corpus vocabulary. Each column of the DTM corresponds to an element of the vocabulary.
  8. Convert an article’s word list into a vector of counts for each term in the vocabulary. This vector is the row of the DTM corresponding to the article.
In [59]:
## I did not have gensim package installed. 
## so the following code is used to install it
## comment it out once it is run once 
#pip install gensim
#pip install wordcloud
#pip install pattern
#pip install textblob
In [60]:
import re
import gensim
from gensim.utils import simple_preprocess
from gensim.models import Phrases  
from gensim.parsing.preprocessing import STOPWORDS
from nltk.stem import WordNetLemmatizer, SnowballStemmer
from nltk.stem.porter import *
#import nltk.stem as stemmer
import numpy as np
from textblob import Word
import nltk
import matplotlib.pyplot as plt
#nltk.download('wordnet')
Matplotlib is building the font cache; this may take a moment.
In [61]:
from wordcloud import WordCloud
In [62]:
## write a function to lemmatize and stem the words 

def remove_email(text):
    return [re.sub('\S*@\S*\s?', 
                   '',
                   email) for email in text]

def lemmatize_stemming(word):
    stemmer =  SnowballStemmer("english")
    #return lemmatize(stemmer.stem(word), pos='v')
    #return Word(word)
    return stemmer.stem(Word(word))

## also remove stop words
def preprocess(text):
    result = []
    for token in gensim.utils.simple_preprocess(text):
        if token not in gensim.parsing.preprocessing.STOPWORDS and len(token)>=3:
            result.append(lemmatize_stemming(token))  
    return result
In [63]:
## apply above functions to an example article 

random_article = np.random.randint(0,100)

one_article = article_data.iloc[random_article]['text']

print('original document: ')
print(one_article)

print('splitted into words: ')
one_artile = remove_email(one_article)
tokens = gensim.utils.simple_preprocess(one_article)
#words = []
#for word in one_article.split():
#    words.append(word)
print(tokens[:100])
print('\n\n tokenized and lemmatized document: ')
words_processed = preprocess(one_article)
print(words_processed[:200])
original document: 
       STOCKHOLM -- Felix Kjellberg doesn't play a very convincing megastar. The seemingly modest Swede avoids the spotlight, has no entourage and is uncomfortable with success. But put Mr. Kjellberg in front of a Web-camera and he transforms into PewDiePie, by far YouTube's biggest draw. He has built a base of 27 million subscribers using a decidedly unorthodox approach to playing video and mobile games. His videos aren't traditional game reviews. "Pewds," as he is often called, simply plays games and allows his audience -- mostly teenagers -- to peer in on his experience and hear random opinions interspersed with odd behavior. He contorts, screeches, swears, sings and even "twerks" to portray his feelings. The 24-year-old Mr. Kjellberg, who created PewDiePie five years ago, has parlayed his persona into a brand name that pulls in the equivalent of $4 million in ad sales a year, most of it pure profit. In December 2012, PewDiePie signed on with Maker Studios, a producer of online content that takes a cut of ad sales. Maker Studios -- which counts on PewDiePie as its most important personality -- sold itself to Walt Disney Co. earlier this year in a deal that could be worth close to $1 billion, depending on certain performance targets. His following is so big that even games he criticizes get coveted publicity. Earlier this year, he made a clip headlined "Flappy Bird -- Don't Play This Game," in which he curses a blue streak while he tries to conquer the then-unknown mobile app. Before long, millions of people had downloaded the game. It helped propel "Flappy Bird" and its Vietnamese developer from obscurity into a world-wide sensation. Mr. Kjellberg also is inadvertently helping to shape the industry, as developers have started making games that aren't just fun to play, but also to watch others play on YouTube, like indie horror games. "It's cool to have this kind of influence, but at the same time it's kind of scary," said Mr. Kjellberg, speaking in a rare interview. He routinely turns down media requests, citing a busy schedule that includes publishing multiple clips a day of himself playing obscure games from an apartment south of London. While his subscriber base creates unparalleled reach on Google Inc.'s YouTube, his success reflects broader momentum for the trend in watching other people play videogames. One of his peers is Jordan Maron, a 22-year-old American known as "Captain Sparklez" who has attracted 7.5 million subscribers with videos related to Mojang's popular Minecraft building-block game. Production quality isn't a key selling point. Mr. Kjellberg's creation process is quick, dirty and done mostly solo. "Unlike many professionally produced shows, I think I've established a much closer contact with my viewers, breaking the wall between the viewer and what's behind the screen," he said. "What I and other YouTubers do is a very different thing, it's almost like hanging around and watching your pal play games. My fans care in a different way about what they are watching." YouTube is playing a more integral role in the experience of gamers. Last week, Sony Corp. announced an update to PlayStation 4 that integrates the online video service into the console so clips of game play can be easily shared. Mr. Kjellberg's career took flight while attending college in Gothenburg, on Sweden's southwest coast. He attended few classes, spending more time at home playing games and uploading to YouTube. After dropping out, he started selling hot dogs. To pad the number of views on his channel during those early days he would repeatedly hit F5 on his keyboard to refresh the browser. As for the name PewDiePie, he said that his original YouTube account was for PewDie -- "pew" to sound like a laser gun, and die for death -- but he lost the password and had to create a new account under a new name, so he added Pie. With growth comes a level of attention that makes Mr. Kjellberg uncomfortable. "I'm so central to YouTube now, and that puts me in the spotlight and raises a lot of questions like 'Why is he so big?'" he said. "I'd much rather prefer to have something like 5 million subscribers." Kevin Lin, the chief operating officer of San Francisco-based Twitch TV -- an online community for videogamers with live streams of game competitions -- said PewDiePie's "strong personality and unique character" helped him achieve a level of success that is hard to crack in traditional forms of media. PewDiePie's ability to draw viewers is valuable to game developers, said Anton Westbergh, chief executive of Sweden's Coffee Stain Studios AB. Coffee Stain developed "Goat Simulator," a game that is wildly popular and profitable, but unpolished and buggy. Mr. Westbergh jokingly describes it as "the world's dumbest game." But PewDiePie's willingness to make videos about Goat Simulator legitimized its existence. "Having guys like PewDiePie playing our game has been tremendous marketing," Mr. Westbergh said. "And for us, there have been no costs involved." Coffee Stain Studios didn't try to steer Mr. Kjellberg's opinion, the CEO said. "What we can do, however, is make a game these guys would like." After Mr. Kjellberg played the bare-bones and difficult "Flappy Bird" in January, he uploaded a video that ripped the game. Its creator, Dong Nguyen, was soon earning over $50,000 a day on a game that had taken just a few days to build. Mr. Nguyen eventually pulled the game, saying it was too addictive. "Apparently [Mr. Nguyen] had a hard time dealing with his success, and I feel bad about that," Mr. Kjellberg said. He notes "99% of developers are super happy that I play their games." PewDiePie's foul language and silly antics can be confusing or offensive to the uninitiated. Mr. Kjellberg said, "I just let go and have fun." Jens Orjeheim, 44, has an 11-year-old son who is a big fan of PewDiePie, but he fails to appreciate the appeal. "I think there are things in society that can be seen as contributing to a positive development," he said. "PewDiePie isn't one of them." He is critical of the fact that Mr. Kjellberg makes money from encouraging children to spend more time in front of screens and elevating the importance of videogames. But Vigor Sorman, founder of a YouTubers network in Sweden, said, "PewDiePie is like a cool friend you have and subscribing to him is almost like Skypeing with him -- that's why viewers are such dedicated fans." Credit: By Sven Grundberg and Jens Hansegard  
splitted into words: 
['stockholm', 'felix', 'kjellberg', 'doesn', 'play', 'very', 'convincing', 'megastar', 'the', 'seemingly', 'modest', 'swede', 'avoids', 'the', 'spotlight', 'has', 'no', 'entourage', 'and', 'is', 'uncomfortable', 'with', 'success', 'but', 'put', 'mr', 'kjellberg', 'in', 'front', 'of', 'web', 'camera', 'and', 'he', 'transforms', 'into', 'pewdiepie', 'by', 'far', 'youtube', 'biggest', 'draw', 'he', 'has', 'built', 'base', 'of', 'million', 'subscribers', 'using', 'decidedly', 'unorthodox', 'approach', 'to', 'playing', 'video', 'and', 'mobile', 'games', 'his', 'videos', 'aren', 'traditional', 'game', 'reviews', 'pewds', 'as', 'he', 'is', 'often', 'called', 'simply', 'plays', 'games', 'and', 'allows', 'his', 'audience', 'mostly', 'teenagers', 'to', 'peer', 'in', 'on', 'his', 'experience', 'and', 'hear', 'random', 'opinions', 'interspersed', 'with', 'odd', 'behavior', 'he', 'contorts', 'screeches', 'swears', 'sings', 'and']


 tokenized and lemmatized document: 
['stockholm', 'felix', 'kjellberg', 'play', 'convinc', 'megastar', 'seem', 'modest', 'swede', 'avoid', 'spotlight', 'entourag', 'uncomfort', 'success', 'kjellberg', 'web', 'camera', 'transform', 'pewdiepi', 'far', 'youtub', 'biggest', 'draw', 'built', 'base', 'million', 'subscrib', 'decid', 'unorthodox', 'approach', 'play', 'video', 'mobil', 'game', 'video', 'aren', 'tradit', 'game', 'review', 'pewd', 'call', 'simpli', 'play', 'game', 'allow', 'audienc', 'teenag', 'peer', 'experi', 'hear', 'random', 'opinion', 'interspers', 'odd', 'behavior', 'contort', 'screech', 'swear', 'sing', 'twerk', 'portray', 'feel', 'year', 'old', 'kjellberg', 'creat', 'pewdiepi', 'year', 'ago', 'parlay', 'persona', 'brand', 'pull', 'equival', 'million', 'sale', 'year', 'pure', 'profit', 'decemb', 'pewdiepi', 'sign', 'maker', 'studio', 'produc', 'onlin', 'content', 'take', 'cut', 'sale', 'maker', 'studio', 'count', 'pewdiepi', 'import', 'person', 'sold', 'walt', 'disney', 'earlier', 'year', 'deal', 'worth', 'close', 'billion', 'depend', 'certain', 'perform', 'target', 'follow', 'big', 'game', 'critic', 'covet', 'public', 'earlier', 'year', 'clip', 'headlin', 'flappi', 'bird', 'play', 'game', 'curs', 'blue', 'streak', 'tri', 'conquer', 'unknown', 'mobil', 'app', 'long', 'million', 'peopl', 'download', 'game', 'help', 'propel', 'flappi', 'bird', 'vietnames', 'develop', 'obscur', 'world', 'wide', 'sensat', 'kjellberg', 'inadvert', 'help', 'shape', 'industri', 'develop', 'start', 'make', 'game', 'aren', 'fun', 'play', 'watch', 'play', 'youtub', 'like', 'indi', 'horror', 'game', 'cool', 'kind', 'influenc', 'time', 'kind', 'scari', 'said', 'kjellberg', 'speak', 'rare', 'interview', 'routin', 'turn', 'media', 'request', 'cite', 'busi', 'schedul', 'includ', 'publish', 'multipl', 'clip', 'day', 'play', 'obscur', 'game', 'apart', 'south', 'london', 'subscrib', 'base', 'creat', 'unparallel', 'reach', 'googl']

Generate a wordcloud for an example article

In [64]:
# Create a WordCloud object
wordcloud = WordCloud(background_color="white", 
                      max_words=500, 
                      contour_width=3, 
                      contour_color='steelblue')
# Generate a word cloud
long_string=(" ").join(words_processed)  
wordcloud.generate(long_string)
# Visualize the word cloud
wordcloud.to_image()
Out[64]:
In [65]:
## process all articles

all_tokens = {}  # empty dictionary to save
id_map = {} ## map the id in the dataset to the id in the matrix (some articles are empty)

id_count = -1  
for i in range(len(article_data)):
    #print(i)
    this_article = article_data['text'].iloc[i]
    if type(this_article)==str:
        #print('article '+str(i) + ' works')
        processed_docs = preprocess(this_article)
        all_tokens[i] = processed_docs
        #print(processed_docs[:10])
        id_count+=1
        id_map[i]=id_count 
        #print(id_count)

all_tokens_list = [all_tokens[i] for i in all_tokens.keys()]
In [67]:
print('Here are an example of the preprocessed words from a particular article')

random_article = np.random.randint(0,100)

print(all_tokens_list[random_article])
Here are an example of the preprocessed words from a particular article
['chines', 'internet', 'compani', 'tencent', 'hold', 'open', 'data', 'center', 'silicon', 'valley', 'week', 'expand', 'cloud', 'comput', 'servic', 'american', 'compani', 'complain', 'face', 'grow', 'restrict', 'china', 'wednesday', 'statement', 'tencent', 'second', 'chines', 'compani', 'open', 'center', 'technolog', 'heartland', 'alibaba', 'group', 'hold', 'china', 'largest', 'commerc', 'compani', 'oper', 'data', 'center', 'east', 'coast', 'alibaba', 'tencent', 'boost', 'cloud', 'comput', 'busi', 'seek', 'tap', 'growth', 'chines', 'compani', 'demand', 'comput', 'power', 'oversea', 'cloud', 'platform', 'offer', 'storag', 'comput', 'network', 'resourc', 'help', 'compani', 'grow', 'lower', 'cost', 'data', 'store', 'access', 'internet', 'reduc', 'need', 'site', 'server', 'tencent', 'statement', 'come', 'month', 'group', 'lawmak', 'wrote', 'letter', 'china', 'ambassador', 'cui', 'tiankai', 'washington', 'china', 'limit', 'cloud', 'comput', 'foreign', 'compani', 'beij', 'requir', 'oversea', 'cloud', 'provid', 'form', 'joint', 'ventur', 'oper', 'countri', 'propos', 'requir', 'turn', 'essenti', 'ownership', 'oper', 'chines', 'partner', 'lawmak', 'argu', 'letter', 'view', 'wall', 'street', 'journal', 'result', 'transfer', 'valuabl', 'intellectu', 'properti', 'accord', 'letter', 'access', 'foreign', 'cloud', 'compani', 'market', 'today', 'restrict', 'past', 'said', 'jake', 'parker', 'vice', 'presid', 'china', 'oper', 'china', 'busi', 'council', 'organ', 'repres', 'multin', 'compani', 'oper', 'countri', 'novemb', 'china', 'ministri', 'industri', 'inform', 'technolog', 'draft', 'notic', 'industri', 'regul', 'potenti', 'shift', 'lot', 'oper', 'control', 'chines', 'partner', 'said', 'final', 'version', 'releas', 'said', 'parker', 'base', 'beij', 'miit', 'wasn', 'immedi', 'avail', 'comment', 'china', 'market', 'cloud', 'infrastructur', 'servic', 'rose', 'billion', 'accord', 'industri', 'research', 'idc', 'microsoft', 'corp', 'amazon', 'com', 'provid', 'servic', 'china', 'joint', 'ventur', 'local', 'partner', 'lag', 'alibaba', 'cloud', 'nation', 'despit', 'strong', 'market', 'share', 'alibaba', 'control', 'market', 'microsoft', 'biggest', 'foreign', 'cloud', 'provid', 'asian', 'countri', 'accord', 'idc', 'tencent', 'run', 'wechat', 'china', 'largest', 'social', 'messag', 'platform', 'million', 'month', 'activ', 'user', 'said', 'increas', 'number', 'data', 'center', 'world', 'wide', 'demand', 'rise', 'cloud', 'servic', 'onlin', 'game', 'internet', 'financ', 'web', 'relat', 'industri', 'shenzhen', 'base', 'compani', 'said', 'open', 'center', 'frankfurt', 'moscow', 'mumbai', 'seoul', 'plan', 'expand', 'silicon', 'valley', 'center', 'dan', 'strumpf', 'contribut', 'articl', 'credit', 'liza', 'lin']

Bag of Words in the Dataset

Create a dictionary from ‘processed_docs’ containing the number of times a word appears in the training set.

In [68]:
from gensim.models import Phrases

## this is a training model of bigrams based on the entire corpus
bigram = Phrases(all_tokens_list,
                 min_count=10)
In [69]:
## add bigrams to the dictionary 

for idx in range(len(all_tokens_list)):
    ## apply the trained model to the texts of each document 
    for token in bigram[[all_tokens_list[idx]]]:
        if '_' in token:
            all_tokens_list[idx].append(token)
In [70]:
## get the dictionary over all articles

dictionary = gensim.corpora.Dictionary(all_tokens_list)

## filter terms that show up in less than 0.1 percent of the articles 
nb_docs = len(all_tokens_list)
no_below = int(0.001*nb_docs)

dictionary.filter_extremes(no_below=no_below)
In [71]:
## print some words from the dictionary 


print('Here are some words from the dictionary:')
count = 0
for x in dictionary.values():
    count=count + 1
    if count <=50:
        print(x)
    
Here are some words from the dictionary:
absent
accus
act
ad
add
advantag
agenda
ago
agre
agreement
alien
allow
allud
america
american
anger
angri
angrili
announc
ask
attack
backfir
bad
barn
barri
base
beat
behavior
benefit
berni
best
betray
better
bigger
biggest
brawl
break
bring
bruis
bush
call
came
candid
capit
card
carolina
case
caught
centuri
certainti
In [72]:
## generate a bag of words for each article 
## i.e. count the frequency of each word in the dictionary in each article 
corpus = [dictionary.doc2bow(i) for i in all_tokens_list]

Initialize LDA model

  • The corpus or the document-term matrix to be passed to the model (in our example is called tokens_matrix)
  • Number of Topics: num_topics is the number of topics we want to extract from the corpus.
  • id2word: It is the mapping from word indices to words. Each of the words has an index that is present in the dictionary.
  • Number of Iterations: it is represented by Passes in Python. Another technical word for iterations is ‘epochs’. Passes control how often we want to train the model on the entire corpus for convergence.
  • Chunksize: It is the number of documents to be used in each training chunk. The chunksize controls how many documents can be processed at one time in the training algorithm.

    • Alpha: is the document-topic density
    • Beta: (In Python, this parameter is called ‘eta’): is the topic word density

      • For instance, the higher values of alpha —> the documents will be composed of more topics, and
      • The lower values of alpha —> returns documents with fewer topics.
In [73]:
import time 
In [74]:
start = time.time()
Lda = gensim.models.ldamodel.LdaModel

##########################################
nb_topics = 180
###########################################


ldamodel = Lda(corpus,
               num_topics= nb_topics,
               id2word=dictionary,
               chunksize = 50000,
               alpha= 1, ## or 'symmetric'
               eta=1.5,
               iterations = 400,
               passes=10,
               random_state=2019,
               eval_every=None)


end = time.time()
In [75]:
print('time taken to run the model: is {} '.format(str(end-start)))
time taken to run the model: is 3357.1156072616577 
In [76]:
print('These are the the most common words for each topic')
ldamodel.print_topics(num_words=20)
These are the the most common words for each topic
Out[76]:
[(146,
  '0.033*"job" + 0.033*"school" + 0.026*"univers" + 0.024*"student" + 0.020*"worker" + 0.017*"work" + 0.014*"colleg" + 0.013*"educ" + 0.012*"employ" + 0.008*"program" + 0.008*"say" + 0.008*"graduat" + 0.008*"state" + 0.007*"hire" + 0.007*"employe" + 0.007*"peopl" + 0.007*"wage" + 0.006*"skill" + 0.006*"labor" + 0.006*"professor"'),
 (109,
  '0.039*"appl" + 0.030*"compani" + 0.018*"smartphon" + 0.018*"mobil" + 0.013*"samsung" + 0.013*"phone" + 0.013*"market" + 0.013*"iphon" + 0.012*"devic" + 0.011*"sale" + 0.011*"electron" + 0.010*"busi" + 0.010*"quarter" + 0.009*"corp" + 0.009*"profit" + 0.009*"soni" + 0.009*"billion" + 0.009*"product" + 0.009*"technolog" + 0.008*"nokia"'),
 (12,
  '0.021*"flight" + 0.019*"safeti" + 0.015*"crash" + 0.014*"airport" + 0.013*"pilot" + 0.011*"train" + 0.010*"plane" + 0.009*"passeng" + 0.009*"accid" + 0.009*"air" + 0.007*"investig" + 0.007*"crew" + 0.007*"fli" + 0.006*"hour" + 0.005*"search" + 0.005*"transport" + 0.005*"control" + 0.005*"airlin" + 0.005*"seat" + 0.005*"travel"'),
 (121,
  '0.006*"peopl" + 0.005*"compani" + 0.005*"includ" + 0.005*"busi" + 0.005*"countri" + 0.004*"nation" + 0.004*"million" + 0.004*"market" + 0.003*"state" + 0.003*"recent" + 0.003*"need" + 0.003*"deal" + 0.003*"industri" + 0.003*"work" + 0.003*"plan" + 0.003*"increas" + 0.003*"world" + 0.003*"group" + 0.003*"report" + 0.003*"billion"'),
 (126,
  '0.005*"week" + 0.004*"state" + 0.004*"busi" + 0.004*"peopl" + 0.004*"countri" + 0.004*"compani" + 0.004*"million" + 0.003*"recent" + 0.003*"public" + 0.003*"group" + 0.003*"help" + 0.003*"presid" + 0.003*"accord" + 0.003*"world" + 0.003*"rule" + 0.003*"work" + 0.003*"call" + 0.003*"plan" + 0.003*"includ" + 0.003*"way"'),
 (24,
  '0.007*"compani" + 0.005*"million" + 0.005*"recent" + 0.005*"countri" + 0.004*"group" + 0.004*"week" + 0.004*"peopl" + 0.003*"call" + 0.003*"includ" + 0.003*"world" + 0.003*"accord" + 0.003*"billion" + 0.003*"end" + 0.003*"long" + 0.003*"major" + 0.003*"need" + 0.003*"offic" + 0.003*"expect" + 0.003*"plan" + 0.003*"increas"'),
 (101,
  '0.009*"billion" + 0.006*"compani" + 0.005*"countri" + 0.005*"week" + 0.005*"market" + 0.005*"million" + 0.004*"expect" + 0.004*"deal" + 0.004*"busi" + 0.004*"bank" + 0.004*"end" + 0.004*"includ" + 0.003*"plan" + 0.003*"financi" + 0.003*"peopl" + 0.003*"econom" + 0.003*"group" + 0.003*"economi" + 0.003*"rais" + 0.003*"invest"'),
 (15,
  '0.051*"rate" + 0.031*"fed" + 0.031*"economi" + 0.025*"growth" + 0.023*"inflat" + 0.021*"polici" + 0.018*"bank" + 0.017*"central" + 0.016*"econom" + 0.015*"economist" + 0.009*"unemploy" + 0.009*"monetari" + 0.009*"reserv" + 0.008*"increas" + 0.008*"term" + 0.008*"recoveri" + 0.007*"feder" + 0.007*"expect" + 0.007*"spend" + 0.007*"price"'),
 (140,
  '0.010*"compani" + 0.006*"peopl" + 0.005*"million" + 0.004*"includ" + 0.004*"need" + 0.004*"countri" + 0.004*"busi" + 0.004*"group" + 0.004*"say" + 0.004*"plan" + 0.004*"recent" + 0.003*"billion" + 0.003*"accord" + 0.003*"week" + 0.003*"market" + 0.003*"increas" + 0.003*"high" + 0.003*"come" + 0.003*"nation" + 0.003*"way"'),
 (98,
  '0.028*"japan" + 0.028*"power" + 0.025*"plant" + 0.020*"energi" + 0.018*"water" + 0.015*"electr" + 0.015*"nuclear" + 0.011*"japanes" + 0.010*"climat" + 0.010*"tokyo" + 0.009*"carbon" + 0.009*"coal" + 0.008*"emiss" + 0.008*"solar" + 0.008*"wind" + 0.008*"util" + 0.008*"reactor" + 0.007*"environment" + 0.006*"fuel" + 0.006*"build"'),
 (106,
  '0.005*"countri" + 0.005*"includ" + 0.005*"compani" + 0.004*"peopl" + 0.004*"need" + 0.004*"billion" + 0.003*"intern" + 0.003*"week" + 0.003*"end" + 0.003*"increas" + 0.003*"accord" + 0.003*"recent" + 0.003*"state" + 0.003*"public" + 0.003*"group" + 0.003*"market" + 0.003*"plan" + 0.003*"world" + 0.003*"day" + 0.003*"deal"'),
 (135,
  '0.005*"peopl" + 0.004*"countri" + 0.004*"world" + 0.004*"nation" + 0.004*"work" + 0.004*"plan" + 0.004*"state" + 0.004*"million" + 0.004*"includ" + 0.004*"econom" + 0.003*"day" + 0.003*"say" + 0.003*"market" + 0.003*"recent" + 0.003*"week" + 0.003*"intern" + 0.003*"expect" + 0.003*"offici" + 0.003*"busi" + 0.003*"end"'),
 (9,
  '0.006*"group" + 0.005*"million" + 0.005*"plan" + 0.005*"countri" + 0.004*"compani" + 0.004*"includ" + 0.004*"billion" + 0.004*"peopl" + 0.004*"public" + 0.004*"end" + 0.003*"accord" + 0.003*"increas" + 0.003*"report" + 0.003*"market" + 0.003*"say" + 0.003*"week" + 0.003*"busi" + 0.003*"world" + 0.003*"recent" + 0.003*"come"'),
 (118,
  '0.006*"billion" + 0.006*"market" + 0.005*"compani" + 0.005*"price" + 0.005*"increas" + 0.005*"week" + 0.005*"report" + 0.005*"million" + 0.005*"countri" + 0.004*"peopl" + 0.004*"recent" + 0.004*"expect" + 0.004*"busi" + 0.004*"global" + 0.003*"chief" + 0.003*"accord" + 0.003*"group" + 0.003*"nation" + 0.003*"need" + 0.003*"start"'),
 (67,
  '0.007*"billion" + 0.006*"includ" + 0.005*"peopl" + 0.004*"week" + 0.004*"market" + 0.004*"recent" + 0.004*"compani" + 0.003*"report" + 0.003*"busi" + 0.003*"group" + 0.003*"deal" + 0.003*"nation" + 0.003*"rule" + 0.003*"work" + 0.003*"countri" + 0.003*"base" + 0.003*"million" + 0.003*"come" + 0.003*"need" + 0.003*"plan"'),
 (123,
  '0.087*"oil" + 0.033*"gas" + 0.030*"price" + 0.027*"energi" + 0.017*"compani" + 0.017*"product" + 0.016*"barrel" + 0.013*"produc" + 0.011*"crude" + 0.010*"natur" + 0.009*"drill" + 0.009*"export" + 0.009*"suppli" + 0.007*"shell" + 0.007*"million" + 0.007*"project" + 0.007*"gulf" + 0.007*"output" + 0.007*"billion" + 0.007*"day"'),
 (138,
  '0.011*"hezbollah" + 0.010*"offici" + 0.005*"lebanon" + 0.005*"countri" + 0.005*"lebanes" + 0.005*"tribun" + 0.004*"forc" + 0.004*"bulgaria" + 0.004*"secur" + 0.004*"state" + 0.004*"peopl" + 0.004*"group" + 0.004*"nation" + 0.004*"recent" + 0.004*"intern" + 0.004*"includ" + 0.004*"presid" + 0.003*"polit" + 0.003*"oper" + 0.003*"administr"'),
 (88,
  '0.004*"market" + 0.004*"includ" + 0.004*"compani" + 0.004*"plan" + 0.004*"countri" + 0.004*"say" + 0.004*"recent" + 0.004*"econom" + 0.003*"peopl" + 0.003*"need" + 0.003*"accord" + 0.003*"week" + 0.003*"million" + 0.003*"group" + 0.003*"increas" + 0.003*"end" + 0.003*"billion" + 0.003*"come" + 0.003*"busi" + 0.003*"world"'),
 (169,
  '0.005*"countri" + 0.004*"plan" + 0.004*"market" + 0.004*"say" + 0.004*"week" + 0.004*"nation" + 0.004*"world" + 0.004*"compani" + 0.004*"help" + 0.004*"peopl" + 0.003*"report" + 0.003*"come" + 0.003*"billion" + 0.003*"need" + 0.003*"cost" + 0.003*"european" + 0.003*"end" + 0.003*"busi" + 0.003*"includ" + 0.003*"group"'),
 (86,
  '0.009*"compani" + 0.006*"busi" + 0.006*"million" + 0.006*"billion" + 0.004*"recent" + 0.004*"week" + 0.004*"peopl" + 0.004*"countri" + 0.004*"accord" + 0.004*"come" + 0.004*"includ" + 0.004*"say" + 0.003*"plan" + 0.003*"help" + 0.003*"base" + 0.003*"work" + 0.003*"nation" + 0.003*"chief" + 0.003*"group" + 0.003*"execut"')]
In [77]:
for i in range(0,ldamodel.num_topics-1):
    print(i)
    print(ldamodel.print_topic(i))
0
0.066*"russia" + 0.050*"russian" + 0.028*"ukrain" + 0.023*"putin" + 0.021*"moscow" + 0.010*"presid" + 0.009*"kremlin" + 0.009*"ukrainian" + 0.009*"soviet" + 0.008*"western"
1
0.007*"peopl" + 0.006*"countri" + 0.005*"nation" + 0.005*"state" + 0.005*"offici" + 0.004*"world" + 0.004*"work" + 0.004*"presid" + 0.004*"group" + 0.004*"week"
2
0.054*"french" + 0.053*"franc" + 0.017*"holland" + 0.016*"pari" + 0.013*"sarkozi" + 0.011*"socialist" + 0.010*"parti" + 0.009*"pen" + 0.009*"presid" + 0.008*"nation"
3
0.006*"countri" + 0.005*"million" + 0.005*"market" + 0.005*"compani" + 0.004*"week" + 0.004*"includ" + 0.004*"billion" + 0.004*"world" + 0.003*"busi" + 0.003*"come"
4
0.012*"oracl" + 0.010*"compani" + 0.010*"cohen" + 0.009*"sap" + 0.008*"sac" + 0.007*"board" + 0.006*"firm" + 0.006*"peopl" + 0.006*"billion" + 0.006*"execut"
5
0.124*"tax" + 0.025*"incom" + 0.019*"cut" + 0.019*"budget" + 0.018*"spend" + 0.016*"rate" + 0.012*"deficit" + 0.010*"pay" + 0.010*"propos" + 0.010*"corpor"
6
0.006*"billion" + 0.006*"million" + 0.005*"plan" + 0.005*"compani" + 0.005*"countri" + 0.004*"report" + 0.004*"market" + 0.004*"week" + 0.004*"includ" + 0.004*"greenberg"
7
0.071*"insur" + 0.017*"luxuri" + 0.010*"life" + 0.009*"loui" + 0.007*"lvmh" + 0.006*"compani" + 0.006*"watch" + 0.006*"good" + 0.006*"herm" + 0.005*"fashion"
8
0.005*"compani" + 0.005*"million" + 0.005*"come" + 0.005*"peopl" + 0.005*"plan" + 0.004*"billion" + 0.004*"market" + 0.004*"includ" + 0.004*"work" + 0.004*"need"
9
0.006*"group" + 0.005*"million" + 0.005*"plan" + 0.005*"countri" + 0.004*"compani" + 0.004*"includ" + 0.004*"billion" + 0.004*"peopl" + 0.004*"public" + 0.004*"end"
10
0.029*"mortgag" + 0.027*"loan" + 0.024*"home" + 0.020*"properti" + 0.020*"hous" + 0.019*"market" + 0.017*"estat" + 0.017*"borrow" + 0.016*"bank" + 0.013*"real"
11
0.006*"million" + 0.005*"billion" + 0.005*"countri" + 0.004*"report" + 0.004*"compani" + 0.004*"includ" + 0.004*"plan" + 0.004*"busi" + 0.004*"state" + 0.004*"need"
12
0.021*"flight" + 0.019*"safeti" + 0.015*"crash" + 0.014*"airport" + 0.013*"pilot" + 0.011*"train" + 0.010*"plane" + 0.009*"passeng" + 0.009*"accid" + 0.009*"air"
13
0.005*"peopl" + 0.005*"recent" + 0.004*"say" + 0.004*"compani" + 0.004*"market" + 0.004*"countri" + 0.004*"million" + 0.004*"plan" + 0.003*"billion" + 0.003*"busi"
14
0.014*"global" + 0.006*"corzin" + 0.005*"peopl" + 0.005*"week" + 0.005*"includ" + 0.004*"compani" + 0.004*"european" + 0.004*"million" + 0.004*"plan" + 0.003*"accord"
15
0.051*"rate" + 0.031*"fed" + 0.031*"economi" + 0.025*"growth" + 0.023*"inflat" + 0.021*"polici" + 0.018*"bank" + 0.017*"central" + 0.016*"econom" + 0.015*"economist"
16
0.042*"africa" + 0.027*"african" + 0.020*"abe" + 0.020*"south" + 0.011*"presid" + 0.009*"minist" + 0.008*"countri" + 0.008*"nation" + 0.007*"parti" + 0.006*"leader"
17
0.006*"market" + 0.006*"plan" + 0.006*"billion" + 0.005*"compani" + 0.005*"recent" + 0.005*"busi" + 0.004*"countri" + 0.004*"week" + 0.004*"expect" + 0.004*"includ"
18
0.042*"airlin" + 0.022*"air" + 0.021*"boe" + 0.020*"carrier" + 0.020*"plane" + 0.017*"jet" + 0.016*"aircraft" + 0.015*"airbus" + 0.011*"unit" + 0.010*"order"
19
0.070*"compani" + 0.033*"share" + 0.029*"deal" + 0.023*"sharehold" + 0.017*"investor" + 0.017*"billion" + 0.015*"firm" + 0.015*"offer" + 0.013*"stake" + 0.013*"stock"
20
0.005*"countri" + 0.005*"busi" + 0.004*"includ" + 0.004*"billion" + 0.004*"market" + 0.004*"increas" + 0.004*"compani" + 0.004*"end" + 0.004*"help" + 0.004*"european"
21
0.032*"drug" + 0.028*"health" + 0.017*"research" + 0.016*"patient" + 0.014*"studi" + 0.013*"care" + 0.013*"medic" + 0.010*"doctor" + 0.010*"hospit" + 0.009*"diseas"
22
0.028*"say" + 0.007*"old" + 0.006*"work" + 0.006*"peopl" + 0.005*"famili" + 0.005*"live" + 0.005*"day" + 0.004*"citi" + 0.004*"home" + 0.004*"film"
23
0.006*"peopl" + 0.005*"compani" + 0.004*"countri" + 0.004*"nation" + 0.004*"includ" + 0.004*"end" + 0.004*"european" + 0.004*"busi" + 0.004*"plan" + 0.004*"recent"
24
0.007*"compani" + 0.005*"million" + 0.005*"recent" + 0.005*"countri" + 0.004*"group" + 0.004*"week" + 0.004*"peopl" + 0.003*"call" + 0.003*"includ" + 0.003*"world"
25
0.013*"compani" + 0.008*"say" + 0.006*"busi" + 0.006*"firm" + 0.006*"peopl" + 0.004*"market" + 0.004*"million" + 0.004*"includ" + 0.004*"way" + 0.004*"accord"
26
0.024*"malaysia" + 0.019*"singapor" + 0.012*"indonesia" + 0.011*"malaysian" + 0.008*"mdb" + 0.008*"najib" + 0.007*"fund" + 0.007*"abu" + 0.007*"money" + 0.006*"countri"
27
0.018*"sweden" + 0.013*"swedish" + 0.007*"countri" + 0.005*"sub" + 0.005*"stockholm" + 0.004*"includ" + 0.004*"thyssenkrupp" + 0.004*"compani" + 0.004*"recent" + 0.004*"say"
28
0.005*"week" + 0.005*"countri" + 0.005*"compani" + 0.005*"plan" + 0.005*"group" + 0.004*"billion" + 0.004*"accord" + 0.004*"million" + 0.004*"recent" + 0.004*"busi"
29
0.089*"north" + 0.079*"south" + 0.068*"korea" + 0.035*"korean" + 0.017*"kim" + 0.013*"seoul" + 0.010*"park" + 0.010*"pyongyang" + 0.009*"missil" + 0.008*"countri"
30
0.005*"compani" + 0.005*"busi" + 0.004*"recent" + 0.004*"week" + 0.004*"come" + 0.004*"countri" + 0.004*"peopl" + 0.003*"billion" + 0.003*"million" + 0.003*"need"
31
0.013*"hay" + 0.008*"compani" + 0.006*"million" + 0.005*"accord" + 0.005*"market" + 0.005*"billion" + 0.004*"busi" + 0.004*"group" + 0.004*"includ" + 0.004*"increas"
32
0.007*"compani" + 0.007*"billion" + 0.006*"plan" + 0.005*"busi" + 0.005*"market" + 0.005*"expect" + 0.005*"includ" + 0.004*"million" + 0.004*"group" + 0.004*"recent"
33
0.033*"protest" + 0.023*"polit" + 0.023*"egypt" + 0.014*"elect" + 0.012*"presid" + 0.012*"egyptian" + 0.010*"democraci" + 0.010*"power" + 0.009*"opposit" + 0.008*"militari"
34
0.014*"india" + 0.008*"compani" + 0.006*"countri" + 0.006*"includ" + 0.005*"million" + 0.004*"market" + 0.004*"busi" + 0.004*"indian" + 0.004*"week" + 0.004*"come"
35
0.007*"compani" + 0.005*"countri" + 0.005*"busi" + 0.004*"group" + 0.004*"accord" + 0.004*"million" + 0.004*"recent" + 0.004*"includ" + 0.004*"market" + 0.004*"billion"
36
0.130*"china" + 0.052*"chines" + 0.023*"yuan" + 0.020*"export" + 0.019*"beij" + 0.018*"currenc" + 0.017*"trade" + 0.014*"economi" + 0.014*"foreign" + 0.012*"growth"
37
0.004*"countri" + 0.004*"peopl" + 0.004*"includ" + 0.004*"compani" + 0.004*"report" + 0.004*"week" + 0.004*"group" + 0.004*"million" + 0.004*"market" + 0.004*"world"
38
0.009*"compani" + 0.007*"avon" + 0.005*"million" + 0.005*"market" + 0.004*"peopl" + 0.004*"repres" + 0.004*"busi" + 0.004*"plan" + 0.004*"product" + 0.004*"includ"
39
0.027*"quarter" + 0.026*"fell" + 0.026*"rose" + 0.021*"index" + 0.018*"stock" + 0.017*"gain" + 0.015*"expect" + 0.013*"share" + 0.012*"market" + 0.011*"point"
40
0.006*"countri" + 0.005*"week" + 0.005*"million" + 0.005*"includ" + 0.004*"state" + 0.004*"peopl" + 0.004*"compani" + 0.004*"help" + 0.004*"end" + 0.004*"say"
41
0.011*"market" + 0.007*"million" + 0.006*"billion" + 0.006*"busi" + 0.005*"compani" + 0.005*"financi" + 0.004*"report" + 0.004*"countri" + 0.004*"week" + 0.004*"expect"
42
0.005*"week" + 0.004*"includ" + 0.004*"peopl" + 0.004*"million" + 0.004*"market" + 0.004*"compani" + 0.004*"plan" + 0.004*"countri" + 0.003*"busi" + 0.003*"presid"
43
0.005*"compani" + 0.005*"countri" + 0.005*"end" + 0.004*"come" + 0.004*"billion" + 0.004*"busi" + 0.004*"world" + 0.004*"includ" + 0.004*"million" + 0.004*"peopl"
44
0.005*"countri" + 0.004*"compani" + 0.004*"state" + 0.004*"includ" + 0.004*"plan" + 0.004*"week" + 0.004*"day" + 0.004*"peopl" + 0.004*"nation" + 0.004*"major"
45
0.005*"peopl" + 0.005*"presid" + 0.004*"countri" + 0.004*"compani" + 0.004*"khodorkovski" + 0.004*"pwc" + 0.004*"rule" + 0.004*"day" + 0.004*"week" + 0.004*"right"
46
0.009*"state" + 0.008*"countri" + 0.007*"presid" + 0.005*"includ" + 0.004*"nation" + 0.004*"week" + 0.004*"leader" + 0.004*"want" + 0.004*"group" + 0.004*"peopl"
47
0.006*"compani" + 0.004*"plan" + 0.004*"market" + 0.004*"million" + 0.004*"group" + 0.004*"includ" + 0.004*"billion" + 0.004*"peopl" + 0.004*"busi" + 0.004*"come"
48
0.005*"compani" + 0.005*"busi" + 0.004*"market" + 0.004*"week" + 0.004*"billion" + 0.004*"includ" + 0.004*"increas" + 0.004*"plan" + 0.004*"million" + 0.004*"accord"
49
0.010*"compani" + 0.006*"busi" + 0.005*"market" + 0.004*"billion" + 0.004*"high" + 0.004*"plan" + 0.004*"recent" + 0.004*"expect" + 0.004*"includ" + 0.004*"say"
50
0.006*"compani" + 0.005*"peopl" + 0.004*"group" + 0.004*"cantor" + 0.004*"week" + 0.004*"accord" + 0.004*"report" + 0.003*"plan" + 0.003*"million" + 0.003*"market"
51
0.005*"countri" + 0.004*"million" + 0.004*"nation" + 0.004*"includ" + 0.004*"come" + 0.004*"recent" + 0.004*"compani" + 0.004*"peopl" + 0.004*"week" + 0.004*"accord"
52
0.026*"compani" + 0.019*"deal" + 0.015*"billion" + 0.011*"valeant" + 0.010*"pharmaceut" + 0.007*"pfizer" + 0.007*"busi" + 0.007*"share" + 0.006*"takeov" + 0.006*"price"
53
0.007*"compani" + 0.006*"billion" + 0.005*"countri" + 0.005*"market" + 0.004*"plan" + 0.004*"increas" + 0.004*"includ" + 0.004*"group" + 0.004*"busi" + 0.004*"million"
54
0.020*"philippin" + 0.007*"countri" + 0.007*"dutert" + 0.006*"manila" + 0.006*"presid" + 0.004*"nation" + 0.004*"peopl" + 0.004*"aquino" + 0.003*"includ" + 0.003*"week"
55
0.008*"billion" + 0.005*"countri" + 0.004*"expect" + 0.004*"market" + 0.004*"group" + 0.004*"world" + 0.004*"nation" + 0.004*"plan" + 0.004*"report" + 0.004*"accord"
56
0.029*"technolog" + 0.023*"space" + 0.020*"chip" + 0.019*"compani" + 0.011*"intel" + 0.010*"tesla" + 0.009*"launch" + 0.008*"satellit" + 0.007*"robot" + 0.007*"rocket"
57
0.008*"compani" + 0.007*"billion" + 0.005*"increas" + 0.005*"plan" + 0.005*"million" + 0.004*"market" + 0.004*"invest" + 0.004*"busi" + 0.004*"group" + 0.004*"come"
58
0.006*"countri" + 0.006*"market" + 0.006*"compani" + 0.005*"expect" + 0.005*"includ" + 0.004*"billion" + 0.004*"peopl" + 0.004*"recent" + 0.004*"week" + 0.004*"busi"
59
0.009*"billion" + 0.008*"compani" + 0.007*"plan" + 0.006*"million" + 0.006*"market" + 0.006*"busi" + 0.004*"includ" + 0.004*"week" + 0.004*"continu" + 0.004*"group"
60
0.005*"state" + 0.004*"compani" + 0.004*"plan" + 0.004*"countri" + 0.004*"peopl" + 0.004*"right" + 0.003*"law" + 0.003*"includ" + 0.003*"come" + 0.003*"need"
61
0.049*"court" + 0.030*"case" + 0.019*"lawyer" + 0.018*"alleg" + 0.018*"judg" + 0.017*"prosecutor" + 0.017*"charg" + 0.013*"trial" + 0.011*"law" + 0.010*"justic"
62
0.007*"countri" + 0.005*"nation" + 0.004*"offici" + 0.004*"week" + 0.004*"group" + 0.004*"state" + 0.004*"say" + 0.004*"includ" + 0.004*"world" + 0.004*"peopl"
63
0.051*"euro" + 0.027*"european" + 0.025*"greec" + 0.021*"zone" + 0.020*"debt" + 0.018*"countri" + 0.016*"germani" + 0.014*"europ" + 0.013*"greek" + 0.013*"german"
64
0.069*"iran" + 0.039*"nuclear" + 0.028*"sanction" + 0.025*"iranian" + 0.016*"weapon" + 0.014*"obama" + 0.014*"deal" + 0.014*"tehran" + 0.013*"offici" + 0.011*"agreement"
65
0.005*"week" + 0.005*"includ" + 0.004*"countri" + 0.004*"accord" + 0.004*"compani" + 0.004*"plan" + 0.003*"group" + 0.003*"chang" + 0.003*"peopl" + 0.003*"recent"
66
0.082*"hong" + 0.080*"kong" + 0.015*"citi" + 0.013*"mainland" + 0.011*"beij" + 0.009*"list" + 0.009*"protest" + 0.008*"china" + 0.008*"asia" + 0.007*"chines"
67
0.007*"billion" + 0.006*"includ" + 0.005*"peopl" + 0.004*"week" + 0.004*"market" + 0.004*"recent" + 0.004*"compani" + 0.003*"report" + 0.003*"busi" + 0.003*"group"
68
0.022*"compani" + 0.020*"market" + 0.017*"brand" + 0.015*"beer" + 0.014*"billion" + 0.014*"sale" + 0.011*"drink" + 0.009*"million" + 0.008*"volum" + 0.007*"inbev"
69
0.062*"news" + 0.021*"corp" + 0.015*"newspap" + 0.014*"media" + 0.012*"report" + 0.012*"hack" + 0.010*"phone" + 0.009*"paper" + 0.008*"broadcast" + 0.008*"alleg"
70
0.008*"market" + 0.007*"compani" + 0.006*"countri" + 0.005*"busi" + 0.005*"week" + 0.004*"plan" + 0.004*"billion" + 0.004*"group" + 0.004*"peopl" + 0.004*"world"
71
0.026*"republican" + 0.026*"trump" + 0.025*"democrat" + 0.025*"obama" + 0.020*"elect" + 0.020*"parti" + 0.019*"presid" + 0.016*"campaign" + 0.015*"vote" + 0.015*"hous"
72
0.089*"trade" + 0.043*"exchang" + 0.024*"market" + 0.019*"firm" + 0.018*"regul" + 0.018*"sec" + 0.017*"stock" + 0.016*"trader" + 0.009*"rule" + 0.009*"commiss"
73
0.006*"compani" + 0.006*"countri" + 0.004*"market" + 0.004*"accord" + 0.004*"world" + 0.004*"report" + 0.004*"plan" + 0.004*"million" + 0.004*"busi" + 0.004*"billion"
74
0.006*"countri" + 0.006*"billion" + 0.006*"compani" + 0.004*"million" + 0.004*"group" + 0.004*"plan" + 0.004*"week" + 0.004*"need" + 0.004*"busi" + 0.003*"expect"
75
0.005*"includ" + 0.005*"countri" + 0.005*"peopl" + 0.005*"come" + 0.004*"report" + 0.004*"market" + 0.004*"million" + 0.004*"work" + 0.004*"accord" + 0.004*"offici"
76
0.010*"bbc" + 0.007*"nation" + 0.005*"countri" + 0.004*"peopl" + 0.004*"intern" + 0.004*"includ" + 0.004*"week" + 0.003*"offici" + 0.003*"come" + 0.003*"group"
77
0.050*"gold" + 0.040*"mine" + 0.029*"metal" + 0.027*"price" + 0.021*"miner" + 0.012*"commod" + 0.012*"demand" + 0.011*"copper" + 0.010*"silver" + 0.010*"ton"
78
0.053*"china" + 0.022*"chines" + 0.016*"beij" + 0.014*"japan" + 0.014*"island" + 0.013*"sea" + 0.010*"region" + 0.010*"asia" + 0.009*"militari" + 0.008*"disput"
79
0.055*"women" + 0.016*"men" + 0.010*"femal" + 0.010*"work" + 0.009*"say" + 0.006*"manag" + 0.006*"divers" + 0.006*"woman" + 0.005*"male" + 0.005*"peopl"
80
0.040*"driver" + 0.023*"uber" + 0.020*"citi" + 0.018*"ride" + 0.015*"compani" + 0.015*"servic" + 0.010*"taxi" + 0.009*"hail" + 0.009*"drive" + 0.008*"car"
81
0.007*"peopl" + 0.005*"includ" + 0.005*"report" + 0.005*"group" + 0.004*"state" + 0.004*"compani" + 0.004*"recent" + 0.004*"week" + 0.004*"million" + 0.004*"countri"
82
0.047*"brazil" + 0.020*"brazilian" + 0.008*"rousseff" + 0.007*"paulo" + 0.007*"countri" + 0.006*"presid" + 0.006*"rio" + 0.006*"silva" + 0.005*"sao" + 0.005*"state"
83
0.005*"billion" + 0.005*"compani" + 0.005*"busi" + 0.005*"countri" + 0.004*"plan" + 0.004*"week" + 0.004*"group" + 0.004*"million" + 0.004*"includ" + 0.004*"recent"
84
0.005*"billion" + 0.005*"compani" + 0.005*"includ" + 0.005*"market" + 0.004*"week" + 0.004*"say" + 0.004*"million" + 0.004*"countri" + 0.004*"nation" + 0.004*"busi"
85
0.010*"million" + 0.009*"dairi" + 0.008*"milk" + 0.008*"busi" + 0.007*"compani" + 0.006*"cow" + 0.006*"plan" + 0.006*"billion" + 0.005*"market" + 0.005*"product"
86
0.009*"compani" + 0.006*"busi" + 0.006*"million" + 0.006*"billion" + 0.004*"recent" + 0.004*"week" + 0.004*"peopl" + 0.004*"countri" + 0.004*"accord" + 0.004*"come"
87
0.007*"compani" + 0.005*"report" + 0.005*"countri" + 0.004*"includ" + 0.004*"recent" + 0.004*"nation" + 0.004*"offici" + 0.004*"help" + 0.003*"million" + 0.003*"peopl"
88
0.004*"market" + 0.004*"includ" + 0.004*"compani" + 0.004*"plan" + 0.004*"countri" + 0.004*"say" + 0.004*"recent" + 0.004*"econom" + 0.003*"peopl" + 0.003*"need"
89
0.006*"compani" + 0.005*"billion" + 0.005*"includ" + 0.005*"market" + 0.004*"week" + 0.004*"increas" + 0.004*"million" + 0.004*"come" + 0.004*"expect" + 0.004*"countri"
90
0.005*"compani" + 0.005*"plan" + 0.004*"market" + 0.004*"recent" + 0.004*"world" + 0.004*"peopl" + 0.004*"week" + 0.004*"billion" + 0.003*"million" + 0.003*"countri"
91
0.006*"compani" + 0.005*"peopl" + 0.005*"million" + 0.004*"recent" + 0.004*"busi" + 0.004*"includ" + 0.004*"plan" + 0.004*"need" + 0.004*"report" + 0.004*"week"
92
0.005*"countri" + 0.005*"presid" + 0.005*"state" + 0.005*"plan" + 0.005*"peopl" + 0.004*"rule" + 0.004*"work" + 0.004*"group" + 0.004*"includ" + 0.004*"need"
93
0.009*"million" + 0.008*"compani" + 0.006*"caterpillar" + 0.006*"busi" + 0.005*"includ" + 0.005*"plan" + 0.004*"expect" + 0.004*"industri" + 0.004*"week" + 0.004*"billion"
94
0.007*"million" + 0.006*"market" + 0.006*"plan" + 0.005*"billion" + 0.005*"compani" + 0.004*"accord" + 0.004*"group" + 0.004*"includ" + 0.004*"busi" + 0.004*"countri"
95
0.030*"polic" + 0.019*"attack" + 0.014*"kill" + 0.011*"peopl" + 0.009*"arrest" + 0.008*"author" + 0.007*"prison" + 0.007*"offici" + 0.007*"suspect" + 0.007*"migrant"
96
0.006*"compani" + 0.004*"busi" + 0.004*"accord" + 0.004*"includ" + 0.004*"help" + 0.004*"world" + 0.004*"countri" + 0.004*"work" + 0.004*"come" + 0.004*"market"
97
0.006*"compani" + 0.005*"billion" + 0.005*"countri" + 0.005*"plan" + 0.004*"market" + 0.003*"million" + 0.003*"peopl" + 0.003*"week" + 0.003*"includ" + 0.003*"come"
98
0.028*"japan" + 0.028*"power" + 0.025*"plant" + 0.020*"energi" + 0.018*"water" + 0.015*"electr" + 0.015*"nuclear" + 0.011*"japanes" + 0.010*"climat" + 0.010*"tokyo"
99
0.013*"compani" + 0.006*"market" + 0.006*"billion" + 0.005*"busi" + 0.005*"countri" + 0.004*"includ" + 0.004*"plan" + 0.004*"million" + 0.004*"oper" + 0.004*"week"
100
0.005*"countri" + 0.004*"week" + 0.004*"peopl" + 0.004*"includ" + 0.004*"state" + 0.004*"public" + 0.003*"day" + 0.003*"compani" + 0.003*"million" + 0.003*"busi"
101
0.009*"billion" + 0.006*"compani" + 0.005*"countri" + 0.005*"week" + 0.005*"market" + 0.005*"million" + 0.004*"expect" + 0.004*"deal" + 0.004*"busi" + 0.004*"bank"
102
0.007*"compani" + 0.005*"peopl" + 0.005*"million" + 0.005*"countri" + 0.005*"includ" + 0.004*"plan" + 0.004*"group" + 0.004*"world" + 0.004*"market" + 0.004*"busi"
103
0.006*"plan" + 0.005*"nation" + 0.005*"week" + 0.004*"peopl" + 0.004*"need" + 0.004*"state" + 0.004*"countri" + 0.004*"million" + 0.003*"group" + 0.003*"end"
104
0.009*"cotton" + 0.008*"market" + 0.006*"billion" + 0.006*"busi" + 0.006*"compani" + 0.005*"plan" + 0.005*"countri" + 0.005*"million" + 0.004*"price" + 0.004*"bank"
105
0.004*"world" + 0.004*"compani" + 0.004*"week" + 0.004*"work" + 0.004*"includ" + 0.004*"peopl" + 0.004*"day" + 0.004*"nation" + 0.004*"plan" + 0.004*"countri"
106
0.005*"countri" + 0.005*"includ" + 0.005*"compani" + 0.004*"peopl" + 0.004*"need" + 0.004*"billion" + 0.003*"intern" + 0.003*"week" + 0.003*"end" + 0.003*"increas"
107
0.006*"compani" + 0.005*"billion" + 0.004*"countri" + 0.004*"week" + 0.004*"peopl" + 0.004*"includ" + 0.004*"recent" + 0.004*"million" + 0.003*"plan" + 0.003*"busi"
108
0.031*"lee" + 0.018*"samsung" + 0.010*"elliott" + 0.007*"note" + 0.007*"compani" + 0.006*"conglomer" + 0.005*"includ" + 0.004*"peopl" + 0.004*"group" + 0.004*"world"
109
0.039*"appl" + 0.030*"compani" + 0.018*"smartphon" + 0.018*"mobil" + 0.013*"samsung" + 0.013*"phone" + 0.013*"market" + 0.013*"iphon" + 0.012*"devic" + 0.011*"sale"
110
0.011*"geithner" + 0.006*"offici" + 0.005*"countri" + 0.005*"econom" + 0.004*"week" + 0.004*"peopl" + 0.004*"nation" + 0.004*"plan" + 0.004*"economi" + 0.003*"meet"
111
0.023*"googl" + 0.017*"facebook" + 0.017*"compani" + 0.016*"internet" + 0.016*"user" + 0.013*"servic" + 0.013*"data" + 0.011*"app" + 0.010*"onlin" + 0.010*"network"
112
0.021*"gambl" + 0.016*"casino" + 0.014*"resort" + 0.012*"vega" + 0.012*"sand" + 0.011*"las" + 0.011*"compani" + 0.008*"macau" + 0.008*"million" + 0.007*"oper"
113
0.026*"game" + 0.025*"team" + 0.015*"player" + 0.013*"play" + 0.011*"leagu" + 0.011*"win" + 0.011*"season" + 0.011*"race" + 0.010*"club" + 0.009*"sport"
114
0.030*"recal" + 0.014*"toyota" + 0.012*"problem" + 0.011*"safeti" + 0.011*"bag" + 0.011*"compani" + 0.010*"switch" + 0.010*"defect" + 0.008*"air" + 0.007*"acceler"
115
0.125*"fund" + 0.046*"manag" + 0.045*"invest" + 0.023*"investor" + 0.021*"hedg" + 0.019*"firm" + 0.018*"asset" + 0.015*"billion" + 0.015*"money" + 0.012*"pension"
116
0.007*"peopl" + 0.005*"group" + 0.004*"compani" + 0.004*"need" + 0.004*"includ" + 0.004*"public" + 0.004*"recent" + 0.004*"million" + 0.003*"world" + 0.003*"busi"
117
0.025*"militari" + 0.020*"afghanistan" + 0.019*"offici" + 0.019*"afghan" + 0.019*"pakistan" + 0.018*"forc" + 0.014*"taliban" + 0.012*"troop" + 0.011*"attack" + 0.010*"kill"
118
0.006*"billion" + 0.006*"market" + 0.005*"compani" + 0.005*"price" + 0.005*"increas" + 0.005*"week" + 0.005*"report" + 0.005*"million" + 0.005*"countri" + 0.004*"peopl"
119
0.042*"australia" + 0.028*"australian" + 0.012*"labor" + 0.009*"minist" + 0.008*"sydney" + 0.007*"prime" + 0.007*"nation" + 0.007*"abbott" + 0.006*"parti" + 0.006*"countri"
120
0.006*"compani" + 0.006*"countri" + 0.005*"peopl" + 0.005*"busi" + 0.005*"market" + 0.005*"million" + 0.004*"includ" + 0.004*"group" + 0.004*"plan" + 0.004*"expect"
121
0.006*"peopl" + 0.005*"compani" + 0.005*"includ" + 0.005*"busi" + 0.005*"countri" + 0.004*"nation" + 0.004*"million" + 0.004*"market" + 0.003*"state" + 0.003*"recent"
122
0.008*"compani" + 0.006*"market" + 0.004*"price" + 0.004*"busi" + 0.004*"million" + 0.004*"increas" + 0.004*"includ" + 0.004*"countri" + 0.004*"world" + 0.004*"recent"
123
0.087*"oil" + 0.033*"gas" + 0.030*"price" + 0.027*"energi" + 0.017*"compani" + 0.017*"product" + 0.016*"barrel" + 0.013*"produc" + 0.011*"crude" + 0.010*"natur"
124
0.010*"compani" + 0.006*"billion" + 0.005*"market" + 0.004*"busi" + 0.004*"peopl" + 0.004*"report" + 0.004*"million" + 0.003*"world" + 0.003*"deal" + 0.003*"plan"
125
0.005*"countri" + 0.005*"group" + 0.004*"week" + 0.004*"offici" + 0.004*"nation" + 0.004*"plan" + 0.004*"come" + 0.003*"peopl" + 0.003*"includ" + 0.003*"work"
126
0.005*"week" + 0.004*"state" + 0.004*"busi" + 0.004*"peopl" + 0.004*"countri" + 0.004*"compani" + 0.004*"million" + 0.003*"recent" + 0.003*"public" + 0.003*"group"
127
0.005*"countri" + 0.005*"group" + 0.004*"week" + 0.004*"plan" + 0.004*"public" + 0.004*"market" + 0.004*"world" + 0.004*"come" + 0.003*"accord" + 0.003*"recent"
128
0.036*"immigr" + 0.024*"state" + 0.023*"law" + 0.017*"mexico" + 0.012*"illeg" + 0.011*"visa" + 0.010*"worker" + 0.010*"popul" + 0.009*"border" + 0.008*"mexican"
129
0.007*"compani" + 0.005*"week" + 0.005*"peopl" + 0.004*"help" + 0.004*"countri" + 0.004*"world" + 0.004*"work" + 0.004*"billion" + 0.003*"includ" + 0.003*"market"
130
0.039*"bond" + 0.035*"market" + 0.032*"investor" + 0.021*"yield" + 0.018*"stock" + 0.017*"rate" + 0.013*"debt" + 0.012*"bank" + 0.012*"price" + 0.011*"currenc"
131
0.007*"compani" + 0.004*"market" + 0.004*"report" + 0.004*"countri" + 0.004*"million" + 0.004*"busi" + 0.004*"plan" + 0.004*"state" + 0.004*"accord" + 0.003*"week"
132
0.032*"parti" + 0.023*"britain" + 0.018*"cameron" + 0.016*"british" + 0.015*"minist" + 0.015*"vote" + 0.015*"prime" + 0.011*"parliament" + 0.011*"referendum" + 0.010*"polit"
133
0.008*"compani" + 0.006*"billion" + 0.005*"market" + 0.005*"countri" + 0.004*"busi" + 0.004*"group" + 0.004*"plan" + 0.004*"peopl" + 0.004*"week" + 0.004*"accord"
134
0.079*"bank" + 0.018*"financi" + 0.015*"morgan" + 0.014*"goldman" + 0.014*"execut" + 0.013*"regul" + 0.012*"client" + 0.011*"familiar" + 0.011*"firm" + 0.010*"swiss"
135
0.005*"peopl" + 0.004*"countri" + 0.004*"world" + 0.004*"nation" + 0.004*"work" + 0.004*"plan" + 0.004*"state" + 0.004*"million" + 0.004*"includ" + 0.004*"econom"
136
0.007*"compani" + 0.006*"market" + 0.006*"billion" + 0.005*"countri" + 0.004*"million" + 0.004*"busi" + 0.004*"peopl" + 0.004*"week" + 0.004*"plan" + 0.004*"includ"
137
0.007*"compani" + 0.004*"busi" + 0.004*"market" + 0.004*"plan" + 0.004*"million" + 0.004*"group" + 0.004*"billion" + 0.004*"need" + 0.004*"recent" + 0.004*"peopl"
138
0.011*"hezbollah" + 0.010*"offici" + 0.005*"lebanon" + 0.005*"countri" + 0.005*"lebanes" + 0.005*"tribun" + 0.004*"forc" + 0.004*"bulgaria" + 0.004*"secur" + 0.004*"state"
139
0.012*"strauss" + 0.011*"kahn" + 0.004*"includ" + 0.004*"presid" + 0.004*"compani" + 0.004*"group" + 0.004*"call" + 0.004*"york" + 0.003*"billion" + 0.003*"dominiqu"
140
0.010*"compani" + 0.006*"peopl" + 0.005*"million" + 0.004*"includ" + 0.004*"need" + 0.004*"countri" + 0.004*"busi" + 0.004*"group" + 0.004*"say" + 0.004*"plan"
141
0.006*"compani" + 0.005*"week" + 0.004*"countri" + 0.004*"report" + 0.004*"million" + 0.004*"group" + 0.004*"market" + 0.004*"world" + 0.004*"includ" + 0.004*"plan"
142
0.006*"countri" + 0.005*"nation" + 0.005*"world" + 0.005*"plan" + 0.004*"includ" + 0.004*"state" + 0.004*"power" + 0.004*"week" + 0.004*"accord" + 0.004*"compani"
143
0.005*"group" + 0.005*"market" + 0.005*"busi" + 0.005*"million" + 0.004*"plan" + 0.004*"compani" + 0.004*"recent" + 0.004*"start" + 0.004*"invest" + 0.003*"expect"
144
0.054*"car" + 0.030*"maker" + 0.030*"sale" + 0.028*"auto" + 0.027*"vehicl" + 0.017*"compani" + 0.016*"motor" + 0.012*"profit" + 0.011*"market" + 0.011*"million"
145
0.006*"compani" + 0.005*"say" + 0.005*"peopl" + 0.004*"plan" + 0.004*"includ" + 0.004*"million" + 0.004*"countri" + 0.004*"nation" + 0.004*"base" + 0.004*"come"
146
0.033*"job" + 0.033*"school" + 0.026*"univers" + 0.024*"student" + 0.020*"worker" + 0.017*"work" + 0.014*"colleg" + 0.013*"educ" + 0.012*"employ" + 0.008*"program"
147
0.007*"compani" + 0.005*"billion" + 0.005*"includ" + 0.004*"recent" + 0.004*"accord" + 0.004*"help" + 0.004*"busi" + 0.004*"expect" + 0.004*"come" + 0.004*"say"
148
0.005*"countri" + 0.005*"plan" + 0.004*"offici" + 0.004*"billion" + 0.004*"compani" + 0.004*"peopl" + 0.004*"includ" + 0.004*"need" + 0.004*"world" + 0.004*"nation"
149
0.005*"group" + 0.005*"peopl" + 0.005*"compani" + 0.004*"countri" + 0.004*"state" + 0.004*"report" + 0.004*"includ" + 0.004*"expect" + 0.004*"say" + 0.004*"plan"
150
0.017*"india" + 0.008*"modi" + 0.006*"compani" + 0.006*"indian" + 0.005*"peopl" + 0.005*"countri" + 0.004*"nation" + 0.004*"state" + 0.004*"billion" + 0.004*"increas"
151
0.005*"compani" + 0.005*"market" + 0.005*"financi" + 0.005*"european" + 0.004*"come" + 0.004*"busi" + 0.004*"billion" + 0.004*"peopl" + 0.004*"includ" + 0.004*"countri"
152
0.039*"sale" + 0.038*"retail" + 0.035*"store" + 0.024*"compani" + 0.016*"consum" + 0.012*"custom" + 0.011*"brand" + 0.011*"onlin" + 0.011*"shop" + 0.010*"product"
153
0.025*"investig" + 0.016*"depart" + 0.015*"email" + 0.014*"inform" + 0.011*"document" + 0.011*"justic" + 0.010*"clinton" + 0.010*"offici" + 0.010*"church" + 0.009*"abus"
154
0.016*"bitcoin" + 0.013*"virtual" + 0.012*"currenc" + 0.008*"money" + 0.005*"payment" + 0.005*"transact" + 0.005*"world" + 0.005*"say" + 0.005*"peopl" + 0.004*"compani"
155
0.018*"yen" + 0.013*"market" + 0.013*"japan" + 0.012*"boj" + 0.009*"bank" + 0.007*"expect" + 0.005*"busi" + 0.005*"polici" + 0.005*"compani" + 0.004*"billion"
156
0.006*"compani" + 0.005*"countri" + 0.004*"say" + 0.004*"week" + 0.004*"come" + 0.004*"peopl" + 0.004*"million" + 0.004*"includ" + 0.004*"billion" + 0.004*"market"
157
0.026*"well" + 0.018*"fargo" + 0.008*"custom" + 0.006*"say" + 0.006*"compani" + 0.006*"peopl" + 0.005*"accord" + 0.005*"million" + 0.005*"account" + 0.005*"group"
158
0.008*"market" + 0.006*"rate" + 0.006*"price" + 0.005*"compani" + 0.005*"econom" + 0.005*"busi" + 0.005*"increas" + 0.005*"growth" + 0.004*"billion" + 0.004*"expect"
159
0.006*"market" + 0.006*"billion" + 0.006*"countri" + 0.004*"come" + 0.004*"need" + 0.004*"includ" + 0.004*"million" + 0.004*"peopl" + 0.004*"busi" + 0.004*"compani"
160
0.005*"peopl" + 0.005*"plan" + 0.005*"countri" + 0.005*"includ" + 0.004*"compani" + 0.004*"busi" + 0.004*"recent" + 0.004*"week" + 0.004*"state" + 0.004*"offici"
161
0.005*"compani" + 0.005*"market" + 0.005*"countri" + 0.005*"state" + 0.004*"group" + 0.004*"peopl" + 0.004*"nation" + 0.004*"need" + 0.004*"includ" + 0.003*"plan"
162
0.123*"bank" + 0.035*"billion" + 0.022*"capit" + 0.017*"loan" + 0.014*"asset" + 0.014*"euro" + 0.010*"profit" + 0.010*"lender" + 0.009*"loss" + 0.009*"financi"
163
0.024*"hotel" + 0.020*"farmer" + 0.019*"compani" + 0.016*"agricultur" + 0.014*"crop" + 0.013*"farm" + 0.011*"deal" + 0.010*"price" + 0.009*"siemen" + 0.009*"seed"
164
0.008*"compani" + 0.005*"plan" + 0.005*"billion" + 0.005*"includ" + 0.004*"busi" + 0.004*"market" + 0.004*"increas" + 0.004*"million" + 0.004*"countri" + 0.004*"week"
165
0.005*"group" + 0.004*"report" + 0.004*"includ" + 0.004*"state" + 0.004*"peopl" + 0.004*"nation" + 0.004*"week" + 0.004*"countri" + 0.004*"public" + 0.004*"recent"
166
0.009*"marin" + 0.006*"schmidt" + 0.005*"state" + 0.005*"peopl" + 0.005*"group" + 0.004*"offici" + 0.004*"river" + 0.004*"presid" + 0.004*"work" + 0.003*"world"
167
0.029*"yahoo" + 0.018*"buffett" + 0.017*"berkshir" + 0.013*"billion" + 0.012*"compani" + 0.011*"alibaba" + 0.009*"busi" + 0.009*"warren" + 0.006*"group" + 0.005*"market"
168
0.004*"compani" + 0.004*"week" + 0.004*"countri" + 0.004*"come" + 0.004*"billion" + 0.004*"secur" + 0.004*"recent" + 0.003*"presid" + 0.003*"busi" + 0.003*"million"
169
0.005*"countri" + 0.004*"plan" + 0.004*"market" + 0.004*"say" + 0.004*"week" + 0.004*"nation" + 0.004*"world" + 0.004*"compani" + 0.004*"help" + 0.004*"peopl"
170
0.013*"coffe" + 0.011*"compani" + 0.009*"chocol" + 0.007*"market" + 0.007*"cocoa" + 0.007*"billion" + 0.007*"bean" + 0.006*"price" + 0.005*"ivori" + 0.005*"million"
171
0.008*"compani" + 0.007*"busi" + 0.006*"million" + 0.006*"billion" + 0.006*"countri" + 0.005*"market" + 0.005*"accord" + 0.004*"plan" + 0.004*"includ" + 0.004*"group"
172
0.022*"islam" + 0.021*"syria" + 0.018*"state" + 0.015*"iraq" + 0.015*"forc" + 0.014*"syrian" + 0.013*"rebel" + 0.011*"militari" + 0.010*"offici" + 0.009*"assad"
173
0.005*"includ" + 0.005*"nation" + 0.005*"compani" + 0.004*"peopl" + 0.004*"week" + 0.004*"countri" + 0.004*"increas" + 0.004*"recent" + 0.004*"million" + 0.004*"busi"
174
0.037*"israel" + 0.029*"turkey" + 0.026*"isra" + 0.021*"palestinian" + 0.014*"turkish" + 0.012*"peac" + 0.010*"erdogan" + 0.009*"minist" + 0.009*"hama" + 0.009*"offici"
175
0.006*"plan" + 0.005*"countri" + 0.005*"includ" + 0.005*"billion" + 0.004*"market" + 0.004*"peopl" + 0.004*"week" + 0.004*"world" + 0.004*"nation" + 0.004*"recent"
176
0.005*"billion" + 0.005*"compani" + 0.005*"includ" + 0.004*"busi" + 0.004*"world" + 0.004*"market" + 0.004*"accord" + 0.004*"million" + 0.004*"countri" + 0.003*"week"
177
0.006*"compani" + 0.005*"includ" + 0.004*"report" + 0.004*"peopl" + 0.004*"market" + 0.004*"world" + 0.004*"accord" + 0.004*"plan" + 0.004*"countri" + 0.004*"recent"
178
0.005*"busi" + 0.004*"includ" + 0.004*"plan" + 0.004*"compani" + 0.004*"peopl" + 0.004*"countri" + 0.004*"billion" + 0.004*"million" + 0.004*"market" + 0.003*"intern"

Assign topics to each article

In [78]:
count = 0 
for i in ldamodel[corpus]:
    if count <=30:
        print('article',count,i)
    count +=1 
article 0 [(71, 0.47222987)]
article 1 [(2, 0.04705763), (22, 0.093155235), (33, 0.031064468), (71, 0.017454904), (95, 0.0115658175), (132, 0.0113791395)]
article 2 [(27, 0.028492287), (63, 0.065099955), (95, 0.08431271), (128, 0.085036606), (153, 0.010551214), (172, 0.011937418)]
article 3 [(7, 0.016200919), (123, 0.062055863)]
article 4 [(63, 0.29349405), (64, 0.010614497)]
article 5 [(5, 0.011643498), (7, 0.010599968), (63, 0.122145005), (66, 0.014367845), (109, 0.06839495), (112, 0.13306877), (123, 0.030898053), (130, 0.11889846)]
article 6 [(2, 0.015218782), (15, 0.109731905), (39, 0.24014212), (130, 0.028926184)]
article 7 [(21, 0.010202953), (33, 0.010451478), (95, 0.04186125), (172, 0.29595888), (174, 0.15965903)]
article 8 [(10, 0.048943464), (72, 0.034056615), (134, 0.016822634), (157, 0.010295764), (162, 0.23267582)]
article 9 [(18, 0.011217641), (19, 0.08878549), (36, 0.039662424), (52, 0.011934907), (66, 0.043953985), (68, 0.010450773), (78, 0.017785162), (113, 0.012966676), (123, 0.04885683), (166, 0.011034027)]
article 10 [(33, 0.015494574), (64, 0.012350334), (132, 0.030704066), (174, 0.084676936)]
article 11 [(22, 0.04320134), (33, 0.012621884), (71, 0.0241439), (128, 0.014395783), (146, 0.012823728), (153, 0.027452363)]
article 12 [(5, 0.049003467), (63, 0.06429873), (64, 0.010571958), (134, 0.21509404), (153, 0.02027469)]
article 13 [(22, 0.12968507), (27, 0.02197051), (109, 0.025597613), (111, 0.17179744), (113, 0.115230165), (170, 0.010930433)]
article 14 [(22, 0.026311086), (56, 0.013300217), (113, 0.43811858), (128, 0.024824126), (146, 0.027796034)]
article 15 [(5, 0.010208345), (19, 0.017190091), (21, 0.029084276), (22, 0.26535922), (71, 0.03063322), (80, 0.011298596), (95, 0.022512913), (146, 0.010776599), (166, 0.01556052)]
article 16 [(39, 0.0703808), (144, 0.1742043), (162, 0.01784714)]
article 17 [(29, 0.0149628045), (64, 0.30221274)]
article 18 [(71, 0.11099465), (117, 0.252005)]
article 19 [(72, 0.033210997), (115, 0.013457512), (134, 0.34521583)]
article 20 [(10, 0.17946568), (61, 0.059083145), (130, 0.040539064), (132, 0.011079536), (134, 0.09821737), (153, 0.013595451), (162, 0.014650425)]
article 21 [(15, 0.026097331), (79, 0.016766125), (109, 0.011948698), (128, 0.027368374), (139, 0.010167529), (154, 0.011775006)]
article 22 [(2, 0.012931575), (19, 0.0106508), (109, 0.03293978), (130, 0.264247), (134, 0.03419273)]
article 23 [(33, 0.3683492), (95, 0.01498651), (117, 0.023945378)]
article 24 [(29, 0.024066271), (64, 0.14829572), (113, 0.012984067)]
article 25 [(0, 0.013374842), (12, 0.012800954), (69, 0.012788942), (78, 0.0321674), (117, 0.059867397), (132, 0.049291354), (174, 0.055991713)]
article 26 [(29, 0.027079908), (36, 0.013913533), (144, 0.23041768)]
article 27 [(69, 0.037060264), (111, 0.013076696), (117, 0.029647363), (166, 0.010321546), (172, 0.23747502)]
article 28 [(10, 0.14269458), (19, 0.110298954), (22, 0.01850817), (63, 0.012794935), (115, 0.067739785), (162, 0.16510333)]
article 29 [(12, 0.014496588), (21, 0.012566338), (22, 0.044779327), (33, 0.056793008), (34, 0.019334486), (79, 0.016460868), (80, 0.016279988), (95, 0.13162458), (128, 0.02973921), (132, 0.010511445), (146, 0.029309122), (150, 0.06475762)]
article 30 [(18, 0.2618246), (64, 0.13246101), (71, 0.025268529)]
  • For instance, article 1 has the highest weight (0.8411557) on the second topic

What is the dominant topic and its percentage contribution in each article?

In [79]:
## a function  
def format_topics_sentences(ldamodel=None, 
                            corpus=corpus, 
                            texts=all_tokens):
    # Init output
    sent_topics_df = pd.DataFrame()

    # Get main topic in each document
    for i, row_list in enumerate(ldamodel[corpus]):
        row = row_list[0] if ldamodel.per_word_topics else row_list            
        # print(row)
        row = sorted(row, key=lambda x: (x[1]), reverse=True)
        # Get the Dominant topic, Perc Contribution and Keywords for each document
        for j, (topic_num, prop_topic) in enumerate(row):
            if j == 0:  # => dominant topic
                wp = ldamodel.show_topic(topic_num)
                topic_keywords = ", ".join([word for word, prop in wp])
                sent_topics_df = sent_topics_df.append(pd.Series([int(topic_num), round(prop_topic,4), topic_keywords]), ignore_index=True)
            else:
                break
    sent_topics_df.columns = ['Dominant_Topic', 'Perc_Contribution', 'Topic_Keywords']

    # Add original text to the end of the output
    contents = pd.Series(texts)
    sent_topics_df = pd.concat([sent_topics_df, contents], axis=1)
    return(sent_topics_df)
In [80]:
df_topic_sents_keywords = format_topics_sentences(ldamodel=ldamodel, corpus=corpus, texts=all_tokens)

# Format
df_dominant_topic = df_topic_sents_keywords.reset_index()
df_dominant_topic.columns = ['Document_No', 'Dominant_Topic', 'Topic_Perc_Contrib', 'Keywords', 'Text']
df_dominant_topic.head(10)
Out[80]:
Document_No Dominant_Topic Topic_Perc_Contrib Keywords Text
0 0 71.0 0.4722 republican, trump, democrat, obama, elect, par... [cnn, exit, poll, new, hampshir, primari, ask,...
1 1 22.0 0.0932 say, old, work, peopl, famili, live, day, citi... [belgium, royal, palac, rais, eyebrow, creat, ...
2 2 128.0 0.0850 immigr, state, law, mexico, illeg, visa, worke... [enjoy, travel, easili, europ, know, long, nee...
3 3 123.0 0.0620 oil, gas, price, energi, compani, product, bar... [nov, editori, america, energi, note, opportun...
4 4 63.0 0.2935 euro, european, greec, zone, debt, countri, ge... [athen, negoti, secur, bailout, deal, time, pr...
5 5 112.0 0.1331 gambl, casino, resort, vega, sand, las, compan... [bet, earli, euro, debt, default, polit, risk,...
6 6 39.0 0.2402 quarter, fell, rose, index, stock, gain, expec... [blue, chip, stock, gave, afternoon, gain, fin...
7 7 172.0 0.2961 islam, syria, state, iraq, forc, syrian, rebel... [suruc, turkey, turkey, push, end, decad, kurd...
8 8 162.0 0.2327 bank, billion, capit, loan, asset, euro, profi... [big, bank, show, sign, recov, financi, crisi,...
9 9 19.0 0.0887 compani, share, deal, sharehold, investor, bil... [chines, conglomer, cosco, group, talk, acquir...
In [81]:
df_dominant_topic.head(30)
Out[81]:
Document_No Dominant_Topic Topic_Perc_Contrib Keywords Text
0 0 71.0 0.4722 republican, trump, democrat, obama, elect, par... [cnn, exit, poll, new, hampshir, primari, ask,...
1 1 22.0 0.0932 say, old, work, peopl, famili, live, day, citi... [belgium, royal, palac, rais, eyebrow, creat, ...
2 2 128.0 0.0850 immigr, state, law, mexico, illeg, visa, worke... [enjoy, travel, easili, europ, know, long, nee...
3 3 123.0 0.0620 oil, gas, price, energi, compani, product, bar... [nov, editori, america, energi, note, opportun...
4 4 63.0 0.2935 euro, european, greec, zone, debt, countri, ge... [athen, negoti, secur, bailout, deal, time, pr...
5 5 112.0 0.1331 gambl, casino, resort, vega, sand, las, compan... [bet, earli, euro, debt, default, polit, risk,...
6 6 39.0 0.2402 quarter, fell, rose, index, stock, gain, expec... [blue, chip, stock, gave, afternoon, gain, fin...
7 7 172.0 0.2961 islam, syria, state, iraq, forc, syrian, rebel... [suruc, turkey, turkey, push, end, decad, kurd...
8 8 162.0 0.2327 bank, billion, capit, loan, asset, euro, profi... [big, bank, show, sign, recov, financi, crisi,...
9 9 19.0 0.0887 compani, share, deal, sharehold, investor, bil... [chines, conglomer, cosco, group, talk, acquir...
10 10 174.0 0.0847 israel, turkey, isra, palestinian, turkish, pe... [bori, johnson, got, disinvit, event, west, ba...
11 11 22.0 0.0431 say, old, work, peopl, famili, live, day, citi... [long, life, michael, novak, travel, write, sp...
12 12 134.0 0.2149 bank, financi, morgan, goldman, execut, regul,... [berlin, german, author, investig, thousand, c...
13 13 111.0 0.1718 googl, facebook, compani, internet, user, serv... [stockholm, felix, kjellberg, play, convinc, m...
14 14 113.0 0.4381 game, team, player, play, leagu, win, season, ... [nfl, draft, time, lie, rumor, laden, run, lea...
15 15 22.0 0.2653 say, old, work, peopl, famili, live, day, citi... [san, diego, summer, famili, arriv, airport, g...
16 16 144.0 0.1743 car, maker, sale, auto, vehicl, compani, motor... [ford, motor, fuel, rise, sale, market, share,...
17 17 64.0 0.3021 iran, nuclear, sanction, iranian, weapon, obam... [unit, nation, panel, expert, said, confidenti...
18 18 117.0 0.2520 militari, afghanistan, offici, afghan, pakista... [washington, gen, david, petraeus, nomin, new,...
19 19 134.0 0.3453 bank, financi, morgan, goldman, execut, regul,... [bank, new, york, mellon, corp, fire, london, ...
20 20 10.0 0.1795 mortgag, loan, home, properti, hous, market, e... [regul, su, morgan, chase, royal, bank, scotla...
21 21 128.0 0.0274 immigr, state, law, mexico, illeg, visa, worke... [editori, polit, war, cash, feb, understand, l...
22 22 130.0 0.2643 bond, market, investor, yield, stock, rate, de... [appl, seek, add, recent, run, blockbust, bond...
23 23 33.0 0.3682 protest, polit, egypt, elect, presid, egyptian... [oust, elect, islamist, presid, juli, egyptian...
24 24 64.0 0.1483 iran, nuclear, sanction, iranian, weapon, obam... [issu, rais, amb, yousef, otaiba, year, iran, ...
25 25 117.0 0.0599 militari, afghanistan, offici, afghan, pakista... [london, turkey, negoti, militari, pact, europ...
26 26 144.0 0.2303 car, maker, sale, auto, vehicl, compani, motor... [seoul, hyundai, motor, affili, kia, motor, co...
27 27 172.0 0.2376 islam, syria, state, iraq, forc, syrian, rebel... [central, intellig, agenc, director, john, bre...
28 28 162.0 0.1651 bank, billion, capit, loan, asset, euro, profi... [balanc, sheet, europ, bank, hold, big, opport...
29 29 95.0 0.1316 polic, attack, kill, peopl, arrest, author, pr... [indian, mourn, death, saturday, year, old, me...
In [82]:
df_dominant_topic.to_pickle('table/dominant_topic.pkl')
In [83]:
"""
print('For instance, the text from an article whose dominant topic is '
      +str(2)
      +str(df_dominant_topic[df_dominant_topic['Dominant_Topic']==2.0]['Text'].iloc[0])[:1000]
     )
"""
Out[83]:
"\nprint('For instance, the text from an article whose dominant topic is '\n      +str(2)\n      +str(df_dominant_topic[df_dominant_topic['Dominant_Topic']==2.0]['Text'].iloc[0])[:1000]\n     )\n"

The most representative sentence for each topic

In [84]:
# Display setting to show more characters in column
pd.options.display.max_colwidth = 100

sent_topics_sorteddf_mallet = pd.DataFrame()
sent_topics_outdf_grpd = df_topic_sents_keywords.groupby('Dominant_Topic')

for i, grp in sent_topics_outdf_grpd:
    sent_topics_sorteddf_mallet = pd.concat([sent_topics_sorteddf_mallet, 
                                             grp.sort_values(['Perc_Contribution'], ascending=False).head(1)], 
                                            axis=0)

# Reset Index    
sent_topics_sorteddf_mallet.reset_index(drop=True, inplace=True)

# Format
sent_topics_sorteddf_mallet.columns = ['Topic_Num', "Topic_Perc_Contrib", "Keywords", "Representative Text"]

# Show
sent_topics_sorteddf_mallet.head(10)

sent_topics_sorteddf_mallet.to_pickle('table/rep_sentence_topic.pkl')

Frequency Distribution of Word Counts in Documents

In [85]:
doc_lens = [len(d) for d in df_dominant_topic.Text if type(d)==list]

# Plot
plt.figure(figsize=(5,3), dpi=160)
plt.hist(doc_lens, bins = 1000, color='navy')
#plt.text(750, 100, "Mean   : " + str(round(np.mean(doc_lens))))
#plt.text(750,  90, "Median : " + str(round(np.median(doc_lens))))
#plt.text(750,  80, "Stdev   : " + str(round(np.std(doc_lens))))
#plt.text(750,  70, "1%ile    : " + str(round(np.quantile(doc_lens, q=0.01))))
#plt.text(750,  60, "99%ile  : " + str(round(np.quantile(doc_lens, q=0.99))))

plt.gca().set(xlim=(0, 1000), ylabel='Number of Documents', xlabel='Document Word Count')
plt.tick_params(size=10)
plt.xticks(np.linspace(0,1000,9))
plt.title('Distribution of Document Word Counts', fontdict=dict(size=22))
plt.savefig('figure/words_distribution.png')

Model evaluations

In [86]:
from gensim.models import CoherenceModel
In [87]:
"""
# Compute Perplexity

print('\nPerplexity: ', ldamodel.log_perplexity(corpus))  
# a measure of how good the model is. lower the better.

# Compute Coherence Score, the higher the better.
coherence_model_lda = CoherenceModel(model=ldamodel, 
                                     texts=all_tokens_list, 
                                     dictionary=dictionary, 
                                     coherence='c_v')

coherence_lda = coherence_model_lda.get_coherence()
print('\nCoherence Score: ', coherence_lda)

"""
Out[87]:
"\n# Compute Perplexity\n\nprint('\nPerplexity: ', ldamodel.log_perplexity(corpus))  \n# a measure of how good the model is. lower the better.\n\n# Compute Coherence Score, the higher the better.\ncoherence_model_lda = CoherenceModel(model=ldamodel, \n                                     texts=all_tokens_list, \n                                     dictionary=dictionary, \n                                     coherence='c_v')\n\ncoherence_lda = coherence_model_lda.get_coherence()\nprint('\nCoherence Score: ', coherence_lda)\n\n"

Visualization of the topic models

In [88]:
## uncomment the code below if there is no pyLDAvis installed 
#pip install pyladvis
In [89]:
import pyLDAvis
import pyLDAvis.gensim_models as gensimvis
In [90]:
pyLDAvis.enable_notebook()
vis = gensimvis.prepare(ldamodel, corpus, dictionary=ldamodel.id2word)
pyLDAvis.save_html(vis, 'figure/WSJfirst_run.html')   ## save it as a html file 
   ## show the figure 
/home/ec2-user/SageMaker/.conda/envs/econnlp/lib/python3.9/site-packages/pyLDAvis/_prepare.py:246: FutureWarning: In a future version of pandas all arguments of DataFrame.drop except for the argument 'labels' will be keyword-only
  default_term_info = default_term_info.sort_values(
/home/ec2-user/SageMaker/.conda/envs/econnlp/lib/python3.9/site-packages/past/builtins/misc.py:45: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses
  from imp import reload
/home/ec2-user/SageMaker/.conda/envs/econnlp/lib/python3.9/site-packages/past/builtins/misc.py:45: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses
  from imp import reload
/home/ec2-user/SageMaker/.conda/envs/econnlp/lib/python3.9/site-packages/past/builtins/misc.py:45: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses
  from imp import reload
/home/ec2-user/SageMaker/.conda/envs/econnlp/lib/python3.9/site-packages/past/builtins/misc.py:45: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses
  from imp import reload
In [76]:
pyLDAvis.display(vis)
Out[76]:

Topic intensity over time

In [ ]:
## in the main dataset, we add columns sized of nb of topics, recording the score of each topic of that article 

for nb in range(nb_topics):
    
    ## for a particular topic 
    weight_dict = {}
    
    for i in range(len(article_data)):
        if i in id_map.keys():
            this_id = id_map[i]  ## id_map maps id in the dataset and in the model 
            weight_list = dict(ldamodel[corpus][this_id])
            
            
            if nb in weight_list:
                #article_data['weight_topic'+str(nb)].iloc[i] = weight_list[nb]
                weight_dict[i] = weight_list[nb]
            else:
                #article_data['weight_topic'+str(nb)].iloc[i] = 0.0
                weight_dict[i] = 0.0
                
    # merge this back to main data
    weight_df = pd.DataFrame(list(weight_dict.items()),
                             columns = ['id','weight_topic'+str(nb)])
    weight_df.set_index('id', drop = True, inplace = True)
    article_data = pd.merge(article_data,
                           weight_df,
                           left_index = True,
                           right_index = True,
                           how='outer')
In [ ]:
## the columns on the right are newly added 

article_data.head(5)
In [ ]:
## The distribution of weight of topic across all articles is 

fig,ax = plt.subplots(1,nb_topics,
                      figsize=(15, 5), 
                      facecolor='w', 
                      edgecolor='k')

for nb in range(nb_topics):
    weights = article_data['weight_topic'+str(nb)]
    weight_min = weights.min()
    weight_max = weights.max()

    print('summary stats:')
    print(weights.describe())

    ## distribution 
    ax[nb].set_title('topic'+str(nb+1))
    ax[nb].hist(weights,
                bins=30)
    ax[nb].set_xlim(weight_min,
             weight_max)
In [ ]:
## day by day 
fig = plt.figure(figsize=(20,5))
for nb in range(nb_topics):
    intensity = article_data.groupby(['date'])['weight_topic'+str(nb)].mean()
    intensity_mv = intensity.rolling(7).mean()
    intensity_mv.plot(lw=3,
                   style='--',
                   label='topic'+str(nb+1))
plt.legend(loc=0)
plt.title('Average Topic Intensity over time (7-day moving average)')
plt.savefig('figure/daily_attention.png')

## notice in this data, the dates are very sparse, hence daily plot may not be very meaningful 
In [ ]:
## month by month 

## day by day 
fig = plt.figure(figsize=(10,5))
for nb in range(nb_topics):
    intensity = article_data.groupby(['month_date'])['weight_topic'+str(nb)].mean()
    #intensity = intensity/intensity[0]
    intensity.plot(lw=4,
                   style='--',
                  label='topic'+str(nb+1))
    #plt.plot(intensity,
    #         lw = 2,
    #         label='topic'+str(nb+1))
plt.legend(loc=0)
plt.title('Average Topic Intensity over time (monthly)')
plt.savefig('figure/monthly_attention.png')
In [ ]:
 
In [ ]: